commit 48e69ee3facb13153f7e89c34c42c509de534d88
Author: CI <toni@data-intuitive.com>
Date:   Mon Aug 11 08:52:14 2025 +0000

    Build branch build/main with version build_main (dfd68e5)
    
    Build pipeline: openpipelines-bio.openpipeline-qc.build-main-fcwdr
    
    Source commit: https://github.com/openpipelines-bio/openpipeline_qc/commit/dfd68e55fafa27fc489f14bfab8ad04315f7f233
    
    Source message: deploy: c7a93716f92762bed306dd7fc40d0b06e6508821

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..79b8508
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,27 @@
+# IDEs and editors
+/.idea
+.project
+.classpath
+*.launch
+.settings/
+.vscode
+
+# Temp
+gitignore
+test_results
+
+# System Files
+.DS_Store
+Thumbs.db
+
+# Nextflow
+work
+.nextflow*
+trace*.txt
+
+# viash
+/resources_test/
+
+# pycache
+*__pycache__*
+
diff --git a/.nojekyll b/.nojekyll
new file mode 100644
index 0000000..e69de29
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000..c12f874
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 openpipelines-bio
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/_viash.yaml b/_viash.yaml
new file mode 100644
index 0000000..8d5bdc5
--- /dev/null
+++ b/_viash.yaml
@@ -0,0 +1,28 @@
+viash_version: 0.9.4
+source: src
+target: target
+name: openpipeline_incubator
+organization: openpipelines-bio
+links:
+  repository: https://github.com/openpipelines-bio/openpipeline_incubator
+  docker_registry: ghcr.io
+repositories:
+  - name: openpipeline
+    repo: openpipelines-bio/openpipeline
+    type: github
+    tag: 2.1.2
+  - name: craftbox
+    repo: craftbox
+    type: vsh
+    tag: v0.2.0
+info:
+  test_resources:
+    - type: s3
+      path: s3://openpipelines-bio/openpipeline_incubator/resources_test
+      dest: resources_test
+config_mods: |
+  .requirements.commands := ['ps']
+  .runners[.type == 'nextflow'].directives.tag := '$id'
+  .resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}
+  .runners[.type == 'nextflow'].config.script := 'includeConfig("nextflow_labels.config")'
+version: build_main
diff --git a/main.nf b/main.nf
new file mode 100644
index 0000000..e69de29
diff --git a/nextflow.config b/nextflow.config
new file mode 100644
index 0000000..e69de29
diff --git a/resources_test_scripts/qc_sample_data.sh b/resources_test_scripts/qc_sample_data.sh
new file mode 100755
index 0000000..ca97c4d
--- /dev/null
+++ b/resources_test_scripts/qc_sample_data.sh
@@ -0,0 +1,166 @@
+#/bin/bash
+
+OUT_DIR=resources_test/qc_sample_data
+OUT_DIR_SPATIAL=resources_test/spatial_qc_sample_data
+
+[ ! -d "$OUT_DIR" ] && mkdir -p "$OUT_DIR"
+[ ! -d "$OUT_DIR_SPATIAL" ] && mkdir -p "$OUT_DIR_SPATIAL"
+
+# fetch/create h5mu from somewhere
+cat > /tmp/params_create_h5mu.yaml <<EOF
+param_list:
+  - id: sample_one
+    input_id: sample_one
+    input: s3://openpipelines-data/10x_5k_anticmv/5k_human_antiCMV_T_TBNK_connect_qc.h5mu
+  - id: sample_two
+    input_id: sample_two
+    input: s3://openpipelines-data/10x_5k_anticmv/5k_human_antiCMV_T_TBNK_connect_qc.h5mu
+output: '\$id.qc.h5mu'
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+# add the sample ID to the mudata object
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/metadata/add_id/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_create_h5mu.yaml \
+  -resume
+
+cat > /tmp/params_subset.yaml <<EOF
+param_list:
+  - id: sample_one
+    input: resources_test/qc_sample_data/sample_one.qc.h5mu
+  - id: sample_two
+    input: resources_test/qc_sample_data/sample_two.qc.h5mu
+output: '\$id.qc.h5mu'
+number_of_observations: 10000
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+# subset h5mus
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/filter/subset_h5mu/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_subset.yaml \
+  -resume
+
+cat > /tmp/add_metadata_obs.py <<EOF
+import mudata as mu
+import glob
+import numpy as np
+import pandas as pd
+import os
+
+# Directory containing the h5mu files
+out_dir = "$(pwd)/resources_test/qc_sample_data"
+
+# List of h5mu files
+h5mu_files = glob.glob(os.path.join(out_dir, "*.h5mu"))
+print(f"Found {len(h5mu_files)} h5mu files: {h5mu_files}")
+
+# Metadata values to randomly assign
+donor_ids = ["donor_1", "donor_2", "donor_3"]
+cell_types = ["CD4+ T cell", "CD8+ T cell", "B cell", "NK cell", "Monocyte"]
+batches = ["batch_A", "batch_B"]
+conditions = ["treated", "control"]
+
+for h5mu_file in h5mu_files:
+    print(f"Processing {h5mu_file}...")
+    
+    # Load MuData object
+    mdata = mu.read_h5mu(h5mu_file)
+    rna = mdata.mod["rna"]
+    n_obs = rna.n_obs
+    
+    # Generate random metadata
+    np.random.seed(42 + hash(h5mu_file) % 100)  # Different seed for each file but reproducible
+    
+    # Create metadata
+    rna.obs["donor_id"] = np.random.choice(donor_ids, size=n_obs)
+    rna.obs["cell_type"] = np.random.choice(cell_types, size=n_obs)
+    rna.obs["batch"] = np.random.choice(batches, size=n_obs)
+    rna.obs["condition"] = np.random.choice(conditions, size=n_obs)
+    
+    # Add a continuous variable too
+    rna.obs["quality_score"] = np.random.uniform(0, 1, size=n_obs)
+    
+    # Save the modified MuData object
+    mu.write_h5mu(h5mu_file, mdata)
+    print(f"Added metadata to {h5mu_file}")
+
+print("All files processed successfully!")
+EOF
+
+# Execute the Python script
+python /tmp/add_metadata_obs.py
+
+# generate cellbender out for testing
+cat > /tmp/params_cellbender.yaml <<EOF
+param_list:
+  - id: sample_one
+    input: resources_test/qc_sample_data/sample_one.qc.h5mu
+  - id: sample_two
+    input: resources_test/qc_sample_data/sample_two.qc.h5mu
+output: '\$id.qc.cellbender.h5mu'
+epochs: 5
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/correction/cellbender_remove_background/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_cellbender.yaml \
+  -resume
+
+# fetch spatial sample data from s3
+aws s3 sync \
+  --profile di \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/spatial_qc_sample_data \
+  "$OUT_DIR_SPATIAL"
+
+# generate json for testing
+viash run src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml --engine docker -- \
+  --input "$OUT_DIR"/sample_one.qc.cellbender.h5mu \
+  --input "$OUT_DIR"/sample_two.qc.cellbender.h5mu \
+  --ingestion_method cellranger_multi \
+  --obs_metadata "donor_id;cell_type;batch;condition" \
+  --output "$OUT_DIR"/sc_dataset.json \
+  --output_reporting_json "$OUT_DIR"/sc_report_structure.json
+
+viash run src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml --engine docker -- \
+  --input "$OUT_DIR_SPATIAL"/xenium_tiny.qc.h5mu \
+  --input "$OUT_DIR_SPATIAL"/xenium_tiny.qc.h5mu \
+  --ingestion_method xenium \
+  --min_num_nonzero_vars 1 \
+  --output "$OUT_DIR_SPATIAL"/xenium_dataset.json \
+  --output_reporting_json "$OUT_DIR_SPATIAL"/xenium_report_structure.json
+
+# remove all state yaml files
+rm "$OUT_DIR"/*.yaml
+rm "$OUT_DIR_SPATIAL"/*.yaml
+
+# copy to s3
+aws s3 sync \
+  "$OUT_DIR" \
+  s3://openpipelines-bio/openpipeline_incubator/"$OUT_DIR" \
+  --delete \
+  --dryrun 
+
+
+aws s3 sync \
+  "$OUT_DIR_SPATIAL" \
+  s3://openpipelines-bio/openpipeline_incubator/"$OUT_DIR_SPATIAL" \
+  --delete \
+  --dryrun 
diff --git a/resources_test_scripts/spatial_qc_sample_data.sh b/resources_test_scripts/spatial_qc_sample_data.sh
new file mode 100755
index 0000000..dca390e
--- /dev/null
+++ b/resources_test_scripts/spatial_qc_sample_data.sh
@@ -0,0 +1,37 @@
+#/bin/bash
+
+OUT_DIR=resources_test/spatial_qc_sample_data
+
+[ ! -d "$OUT_DIR" ] && mkdir -p "$OUT_DIR"
+
+# fetch/create h5mu from somewhere
+cat > /tmp/qc.yaml <<EOF
+param_list:
+  - id: xenium_tiny
+    input: s3://openpipelines-bio/openpipeline_spatial/resources_test/xenium/xenium_tiny.h5mu
+  - id: Lung5_Rep2_tiny
+    input: s3://openpipelines-bio/openpipeline_spatial/resources_test/cosmx/Lung5_Rep2_tiny.h5mu
+var_name_mitochondrial_genes: mitochondrial
+var_name_ribosomal_genes: ribosomal
+output: '\$id.qc.h5mu'
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.0 \
+  -main-script target/nextflow/workflows/qc/qc/main.nf \
+  -profile docker \
+  -params-file /tmp/qc.yaml \
+  -resume \
+  -config src/configs/labels_ci.config
+
+# copy to s3
+aws s3 sync \
+  --profile di \
+  resources_test/spatial_qc_sample_data \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/spatial_qc_sample_data \
+  --delete --dryrun \
+  --exclude "*" --include "*.h5mu" \
+  
diff --git a/src/authors/dorien_roosen.yaml b/src/authors/dorien_roosen.yaml
new file mode 100644
index 0000000..b23fca0
--- /dev/null
+++ b/src/authors/dorien_roosen.yaml
@@ -0,0 +1,11 @@
+name: Dorien Roosen
+info:
+  role: Core Team Member
+  links:
+    email: dorien@data-intuitive.com
+    github: dorien-er
+    linkedin: dorien-roosen
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Data Scientist
diff --git a/src/authors/jakub_majercik.yaml b/src/authors/jakub_majercik.yaml
new file mode 100644
index 0000000..1dac285
--- /dev/null
+++ b/src/authors/jakub_majercik.yaml
@@ -0,0 +1,11 @@
+name: Jakub Majercik
+info:
+  role: Contributor
+  links:
+    email: jakub@data-intuitive.com
+    github: jakubmajercik
+    linkedin: jakubmajercik
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Bioinformatics Engineer
diff --git a/src/authors/robrecht_cannoodt.yaml b/src/authors/robrecht_cannoodt.yaml
new file mode 100644
index 0000000..8f0fe78
--- /dev/null
+++ b/src/authors/robrecht_cannoodt.yaml
@@ -0,0 +1,15 @@
+name: Robrecht Cannoodt
+info:
+  role: Core Team Member
+  links:
+    email: robrecht@data-intuitive.com
+    github: rcannood
+    orcid: "0000-0003-3641-729X"
+    linkedin: robrechtcannoodt
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Data Science Engineer
+    - name: Open Problems
+      href: https://openproblems.bio
+      role: Core Member
diff --git a/src/authors/weiwei_schultz.yaml b/src/authors/weiwei_schultz.yaml
new file mode 100644
index 0000000..04d8904
--- /dev/null
+++ b/src/authors/weiwei_schultz.yaml
@@ -0,0 +1,6 @@
+name: Weiwei Schultz
+info:
+  role: Contributor
+  organizations:
+    - name: Janssen R&D US
+      role: Associate Director Data Sciences
diff --git a/src/base/requirements/anndata.yaml b/src/base/requirements/anndata.yaml
new file mode 100644
index 0000000..60939e6
--- /dev/null
+++ b/src/base/requirements/anndata.yaml
@@ -0,0 +1,2 @@
+packages:
+  - anndata~=0.11.1
\ No newline at end of file
diff --git a/src/base/requirements/anndata_mudata.yaml b/src/base/requirements/anndata_mudata.yaml
new file mode 100644
index 0000000..d324e81
--- /dev/null
+++ b/src/base/requirements/anndata_mudata.yaml
@@ -0,0 +1,9 @@
+__merge__: [/src/base/requirements/anndata.yaml, .]
+packages:
+  - mudata~=0.3.1
+# Make sure that awkward is not installed. Currently, support of awkward arrays
+# in anndata is experimental, and it is enabled based on whether or not the package
+# is available. By making sure that awkward is not installed, the functionality is
+# not enabled.
+script: |
+  exec("try:\n  import awkward\nexcept ModuleNotFoundError:\n  exit(0)\nelse:  exit(1)")
\ No newline at end of file
diff --git a/src/base/requirements/viashpy.yaml b/src/base/requirements/viashpy.yaml
new file mode 100644
index 0000000..a6dcce4
--- /dev/null
+++ b/src/base/requirements/viashpy.yaml
@@ -0,0 +1,2 @@
+packages:
+  - viashpy==0.8.0
\ No newline at end of file
diff --git a/src/configs/integration_tests.config b/src/configs/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/src/configs/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/src/configs/labels.config b/src/configs/labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/src/configs/labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/src/configs/labels_ci.config b/src/configs/labels_ci.config
new file mode 100644
index 0000000..7a1c7d3
--- /dev/null
+++ b/src/configs/labels_ci.config
@@ -0,0 +1,105 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
+
+profiles {
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.fixOwnership    = true
+    docker.enabled         = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+
+  local {
+    // This config is for local processing.
+    process {
+        maxMemory = 25.GB
+        withLabel: verylowcpu { cpus = 2 }
+        withLabel: lowcpu { cpus = 4 }
+        withLabel: midcpu { cpus = 6 }
+        withLabel: highcpu { cpus = 12 }
+  
+        withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+        withLabel: midmem { memory = { get_memory( 12.GB * task.attempt ) } }
+        withLabel: highmem { memory = { get_memory( 20.GB * task.attempt ) } }
+    }
+  }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/src/ingestion_qc/detect_ingestion_method/config.vsh.yaml b/src/ingestion_qc/detect_ingestion_method/config.vsh.yaml
new file mode 100644
index 0000000..cb461be
--- /dev/null
+++ b/src/ingestion_qc/detect_ingestion_method/config.vsh.yaml
@@ -0,0 +1,67 @@
+name: detect_ingestion_method
+namespace: ingestion_qc
+description: | 
+  Detects the ingestion method of a dataset.
+  Currently detects either 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other technologies upon request.
+authors: 
+  - __merge__: /src/authors/dorien_roosen.yaml
+    roles: [author]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [contributor]
+argument_groups:
+  - name: Inputs
+    arguments:
+      - name: --input
+        type: file
+        required: true
+        direction: input
+        description: The input h5mu file(s)
+        example: path/to/file.h5mu
+      - name: --modality
+        type: string
+        description: The modality to use
+        default: rna
+  - name: Outputs
+    arguments:
+      - name: --output_uns_ingestion_method
+        description: The .uns field in which to store the exprimental setup. Values stored are `cellranger_multi`, `xenium` or `cosmx`.
+        type: string
+        default: ingestion_method
+      - name: --output
+        type: file
+        required: true
+        direction: output
+        description: The output h5mu file, containing an .uns field with experiment description.
+        example: path/to/file.h5mu
+resources:
+  - type: python_script
+    path: script.py
+  - path: /src/utils/setup_logger.py
+test_resources:
+  - type: python_script
+    path: test.py
+  - path: /resources_test/qc_sample_data/sample_one.qc.h5mu
+  - path: /resources_test/spatial_qc_sample_data/Lung5_Rep2_tiny.qc.h5mu
+  - path: /resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu
+engines:
+  - type: docker
+    image: python:3.12-slim
+    setup:
+      - type: apt
+        packages:
+          - procps
+      - type: python
+        __merge__: [/src/base/requirements/anndata_mudata.yaml]
+    test_setup:
+    - type: apt
+      packages:
+        - git
+    - type: python
+      __merge__: [/src/base/requirements/viashpy.yaml]
+      github: openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils
+
+runners:
+  - type: executable
+  - type: nextflow
+    directives:
+      label: [lowmem, lowdisk]
diff --git a/src/ingestion_qc/detect_ingestion_method/script.py b/src/ingestion_qc/detect_ingestion_method/script.py
new file mode 100644
index 0000000..df69ae6
--- /dev/null
+++ b/src/ingestion_qc/detect_ingestion_method/script.py
@@ -0,0 +1,69 @@
+import shutil
+import anndata as ad
+import h5py
+import sys
+
+## VIASH START
+par = {
+    "input": "resources_test/qc_sample_data/sample_one.qc.h5mu",
+    # "input": "resources_test/spatial_qc_sample_data/xenium/xenium_tiny_qc.h5mu",
+    # "input": "/Users/dorienroosen/code/openpipeline_spatial/resources_test/cosmx/Lung5_Rep2_tiny.h5mu",
+    "output": "output.h5mu",
+    "output_uns_ingestion_method": "ingestion_method",
+    "modality": "rna"
+}
+meta = {
+    "resources_dir": "src/utils"
+}
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main(par):
+
+    # read h5mu file
+    with h5py.File(par["input"], "r") as file:
+        mod = file["mod"][par["modality"]]
+        uns = ad.experimental.read_elem(file["uns"])
+        mod_obs = ad.experimental.read_elem(mod["obs"])
+        mod_uns = ad.experimental.read_elem(mod["uns"])
+
+    # detect ingestion method
+    ingestion_methods = {
+            "cellranger_multi": "metrics_cellranger" in uns,
+            "xenium": all(key in mod_obs for key in ["segmentation_method", "nucleus_area"]),
+            "cosmx": "spatial" in mod_uns and "fov" in mod_obs
+        }
+
+    # make sure only one ingestion method is detected
+    detected_methods = [method for method, detected in ingestion_methods.items() if detected]
+    methods_count = len(detected_methods)
+
+    if methods_count == 1:
+        detected_method = detected_methods[0]
+        logger.info(f"Detected ingestion method {detected_method}")
+
+    elif methods_count == 0:
+        raise ValueError("No ingestion method detected")
+
+    else:
+        raise ValueError(f"Multiple ingestion methods detected: {', '.join(detected_methods)}")
+
+    # check if mod_uns already contains a different detected method
+    if mod_uns.get(par["output_uns_ingestion_method"], detected_method) != detected_method:
+        raise ValueError(f"Field .uns['{par['output_uns_ingestion_method']}'] already exists and contains different value `{mod_uns.get(par['output_uns_ingestion_method'])}` than detected method (`{detected_method}`).")
+
+    # copy input to output
+    shutil.copy(par["input"], par["output"])
+
+    if par["output_uns_ingestion_method"] not in mod_uns:
+        with h5py.File (par["output"], "r+") as out_file:
+            out_file["uns"][par["output_uns_ingestion_method"]] = detected_method
+
+
+if __name__ == "__main__":
+    main(par)
diff --git a/src/ingestion_qc/detect_ingestion_method/test.py b/src/ingestion_qc/detect_ingestion_method/test.py
new file mode 100644
index 0000000..e24cab9
--- /dev/null
+++ b/src/ingestion_qc/detect_ingestion_method/test.py
@@ -0,0 +1,63 @@
+import pytest
+import h5py
+import os
+import anndata as ad
+import sys
+
+## VIASH START
+meta = {
+    "resources_dir": "resources_test"
+}
+## VIASH END
+
+
+def test_cellranger(run_component, tmp_path):
+    output = tmp_path / "output_cellranger.h5mu"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/sample_one.qc.h5mu",
+            "--output", output
+        ]
+    )
+
+    assert os.path.exists(output), "Output file was not created"
+    with h5py.File (output, "r") as out_file:
+        uns = ad.experimental.read_elem(out_file["uns"])
+        assert uns["ingestion_method"] == "cellranger_multi", "cellranger_multi not detected"
+
+
+def test_xenium(run_component, tmp_path):
+    output = tmp_path / "output_xenium.h5mu"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/xenium_tiny.qc.h5mu",
+            "--output", output
+        ]
+    )
+
+    assert os.path.exists(output), "Output file was not created"
+    with h5py.File (output, "r") as out_file:
+        uns = ad.experimental.read_elem(out_file["uns"])
+        assert uns["ingestion_method"] == "xenium", "xenium not detected"
+
+
+def test_cosmx(run_component, tmp_path):
+    output = tmp_path / "output_cosmx.h5mu"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/Lung5_Rep2_tiny.qc.h5mu",
+            "--output", output
+        ]
+    )
+
+    assert os.path.exists(output), "Output file was not created"
+    with h5py.File (output, "r") as out_file:
+        uns = ad.experimental.read_elem(out_file["uns"])
+        assert uns["ingestion_method"] == "cosmx", "cosmx not detected"
+
+
+if __name__ == "__main__":
+    sys.exit(pytest.main([__file__]))
\ No newline at end of file
diff --git a/src/ingestion_qc/generate_html/config.vsh.yaml b/src/ingestion_qc/generate_html/config.vsh.yaml
new file mode 100644
index 0000000..ab31128
--- /dev/null
+++ b/src/ingestion_qc/generate_html/config.vsh.yaml
@@ -0,0 +1,64 @@
+name: generate_html
+namespace: ingestion_qc
+description: Generate an HTML report from the QC metrics
+authors: 
+  - __merge__: /src/authors/jakub_majercik.yaml
+    roles: [author]
+  - __merge__: /src/authors/dorien_roosen.yaml
+    roles: [author]
+  - __merge__: /src/authors/robrecht_cannoodt.yaml
+    roles: [author]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [contributor]
+argument_groups:
+  - name: Inputs
+    arguments:
+      - name: --input_data
+        type: file
+        required: true
+        direction: input
+        description: The input JSON file containing the QC metrics
+        example: path/to/file.json
+      - name: --input_structure
+        type: file
+        required: true
+        direction: input
+        description: The input JSON file containing the structure of the data
+        example: path/to/file.json
+  - name: Outputs
+    arguments:
+      - name: --output_qc_report
+        type: file
+        required: true
+        direction: output
+        description: The output HTML report
+        example: path/to/file.html
+resources:
+  - type: bash_script
+    path: script.sh
+test_resources:
+  - type: bash_script
+    path: test.sh
+  - path: /resources_test/qc_sample_data/sc_dataset.json
+  - path: /resources_test/qc_sample_data/sc_report_structure.json
+  - path: /resources_test/spatial_qc_sample_data/xenium_dataset.json
+  - path: /resources_test/spatial_qc_sample_data/xenium_report_structure.json
+engines:
+  - type: docker
+    image: node:latest
+    setup:
+      - type: apt
+        packages:
+          - git
+      - type: docker
+        run: |
+          npm install -g pnpm@latest-10 \
+          && cd /opt && git clone https://github.com/openpipelines-bio/incubator_ingestion_qc.git \
+          && cd incubator_ingestion_qc && pnpm install \
+          && true
+
+runners:
+  - type: executable
+  - type: nextflow
+    directives:
+      label: [lowmem, lowdisk]
diff --git a/src/ingestion_qc/generate_html/script.sh b/src/ingestion_qc/generate_html/script.sh
new file mode 100755
index 0000000..4bd9cf8
--- /dev/null
+++ b/src/ingestion_qc/generate_html/script.sh
@@ -0,0 +1,18 @@
+ABSOLUTE_INPUT_DATA=$(realpath $par_input_data)
+ABSOLUTE_INPUT_STRUCTURE=$(realpath $par_input_structure)
+ABSOLUTE_OUTPUT=$(realpath $par_output_qc_report)
+
+cd /opt/incubator_ingestion_qc
+mkdir src/data
+
+echo "Compressing input data..."
+pnpm run compress_data "$ABSOLUTE_INPUT_DATA" "src/data/dataset.ts"
+
+echo "Compressing report structure..."
+pnpm run compress_data "$ABSOLUTE_INPUT_STRUCTURE" "src/data/report_structure.ts"
+
+echo "Generating HTML..."
+pnpm run build
+
+echo "Copying HTML to output directory..."
+cp dist/index.html "$ABSOLUTE_OUTPUT"
diff --git a/src/ingestion_qc/generate_html/test.sh b/src/ingestion_qc/generate_html/test.sh
new file mode 100644
index 0000000..f1d2b24
--- /dev/null
+++ b/src/ingestion_qc/generate_html/test.sh
@@ -0,0 +1,10 @@
+echo ">> Generating report"
+"$meta_executable" \
+    --input_data "$meta_resources_dir/sc_dataset.json" \
+    --input_structure "$meta_resources_dir/sc_report_structure.json" \
+    --output_qc_report "index.html" \
+
+echo ">> Checking output"
+[ ! -f "index.html" ] && echo "Error: Output report does not exist." && exit 1
+
+echo ">> Test succesful" && exit 0
\ No newline at end of file
diff --git a/src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml b/src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml
new file mode 100644
index 0000000..25f1a75
--- /dev/null
+++ b/src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml
@@ -0,0 +1,168 @@
+name: h5mu_to_qc_json
+namespace: ingestion_qc
+scope: private
+description: |
+  Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and processed by the QC workflow, and generates:
+  - A JSON file that contains the combined data for the QC report
+  - A JSON file that defines the layout and structure of the QC report
+authors: 
+  - __merge__: /src/authors/jakub_majercik.yaml
+    roles: [author]
+  - __merge__: /src/authors/dorien_roosen.yaml
+    roles: [author]
+  - __merge__: /src/authors/robrecht_cannoodt.yaml
+    roles: [author]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [contributor]
+argument_groups:
+  - name: Inputs
+    arguments:
+      - name: --input
+        type: file
+        multiple: true
+        required: true
+        direction: input
+        description: The input h5mu file(s)
+        example: path/to/file.h5mu
+      - name: --modality
+        type: string
+        description: The modality to use
+        default: rna
+      - name: --ingestion_method
+        type: string
+        required: true
+        choices:
+          - cellranger_multi
+          - xenium
+        description: Method that was used to ingest the data - this will define the structure of the report that is generated.
+      - name: --obs_sample_id
+        type: string
+        description: The key in the h5mu file that contains the sample ID. If not provided, each H5MU file will be considered as a separate sample.
+        default: sample_id
+      - name: --obs_total_counts
+        type: string
+        description: The key in the h5mu .obs field that contains the total counts.
+        default: total_counts
+      - name: --obs_num_nonzero_vars
+        type: string
+        description: The key in the h5mu .obs field that contains the number of nonzero vars.
+        default: num_nonzero_vars
+      - name: --obs_fraction_mitochondrial
+        type: string
+        description: The key in the h5mu .obs field that contains the fraction mitochondrial genes.
+        default: fraction_mitochondrial
+      - name: --obs_fraction_ribosomal
+        type: string
+        description: The key in the h5mu .obs field that contains the fraction ribosomal genes.
+        default: fraction_ribosomal
+
+  - name: Outputs
+    arguments:
+      - name: --output
+        type: file
+        required: true
+        direction: output
+        description: The output JSON file
+        example: path/to/file.json
+      - name: --output_reporting_json
+        type: file
+        required: true
+        description: The output JSON file that defines the QC report
+        direction: output
+        example: path/to/file.json
+
+  - name: Filtering & grouping options
+    arguments:
+      - name: --min_total_counts
+        type: integer
+        description: Minimum total counts for a cell to be included in the output
+        default: 10
+      - name: --min_num_nonzero_vars
+        type: integer
+        description: Minimum number of nonzero vars for a cell to be included in the output
+        default: 10
+      - name: --obs_metadata
+        type: string
+        multiple: true
+        description: The metadata keys in the h5mu .obs to include in the output JSON.
+        example: "donor_id;cell_type;batch;condition"
+
+  - name: Options for CellRanger reports
+    arguments:
+      - name: --obs_cellbender
+        type: string
+        multiple: true
+        description: The cellbender keys in the h5mu .obs to include in the output JSON
+        default: [
+          "cellbender_background_fraction",
+          "cellbender_cell_probability",
+          "cellbender_cell_size",
+          "cellbender_droplet_efficiency"
+          ]
+      - name: --uns_cellranger_metrics
+        type: string
+        description: The key in the h5mu file .uns that contains the cellranger metrics
+        default: metrics_cellranger
+
+
+  - name: Options for Xenium reports
+    arguments:
+      - name: --obs_nucleus_area
+        type: string
+        description: The key in the h5mu .obs field that contains the nucleus area.
+        default: nucleus_area
+      - name: --obs_cell_area
+        type: string
+        description: The key in the h5mu .obs field that contains the cell area.
+        default: cell_area
+      - name: --obs_x_coord
+        type: string
+        description: The key in the h5mu .obs field that contains the x coordinate.
+        default: x_coord
+      - name: --obs_y_coord
+        type: string
+        description: The key in the h5mu .obs field that contains the y coordinate.
+        default: y_coord
+      - name: --obs_control_probe_counts
+        type: string
+        description: The key in the h5mu .obs field that contains the number of control probes.
+        default: control_probe_counts
+      - name: --obs_control_codeword_counts
+        type: string
+        description: The key in the h5mu .obs field that contains the number of control codewords.
+        default: control_codeword_counts  
+
+  # - name: Options for CosMx reports
+
+resources:
+  - type: python_script
+    path: script.py
+  - path: /src/utils/setup_logger.py
+  - path: report_structure
+test_resources:
+  - type: python_script
+    path: test.py
+  - type: file
+    path: /resources_test
+engines:
+  - type: docker
+    image: python:3.12-slim
+    setup:
+      - type: apt
+        packages:
+          - procps
+      - type: python
+        __merge__: [ /src/base/requirements/anndata_mudata.yaml ]
+    test_setup:
+    - type: apt
+      packages:
+        - git
+    - type: python
+      __merge__: [/src/base/requirements/viashpy.yaml]
+      github: openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils
+
+runners:
+  - type: executable
+  - type: nextflow
+    directives:
+      label: [midmem, middisk]
diff --git a/src/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json b/src/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
new file mode 100644
index 0000000..18769ba
--- /dev/null
+++ b/src/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
@@ -0,0 +1,162 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "SampleQC",
+      "key": "metrics_cellranger_stats",
+      "additionalAxes": false,
+      "defaultFilters": [
+        {
+          "type": "bar",
+          "field": "Number_of_reads_in_the_library",
+          "label": "Number of reads per library",
+          "description": "Sequencing depth per sample. Higher values generally indicate more comprehensive cell profiling.",
+          "nBins": 10,
+          "groupBy": "sample_id",
+          "xAxisType": "linear",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Confidently_mapped_reads_in_cells",
+          "label": "Confidently mapped reads in cells",
+          "description": "Number of reads that were mapped unambiguously to the reference genome within cell-containing droplets.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Estimated_number_of_cells",
+          "label": "Estimated number of cells",
+          "description": "CellRanger's estimate of the number of cells per sample based on the UMI count distribution.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Sequencing_saturation",
+          "label": "Sequencing saturation",
+          "description": "Fraction of reads that are duplicates of existing UMIs. Higher values suggest deeper sequencing coverage.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        }
+      ]
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "pct_of_counts_in_top_50_vars",
+          "label": "Fraction UMI in top 50 genes per cell",
+          "description": "Proportion of RNA molecules from the 50 most-expressed genes in each cell.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_probability",
+          "label": "CellBender cell probability",
+          "description": "CellBender's statistical confidence (0-1) that a barcode represents a real cell, with higher values indicating stronger confidence.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_background_fraction",
+          "label": "CellBender background fraction",
+          "description": "Estimated percentage of each cell's RNA that comes from the ambient solution rather than the cell itself.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_size",
+          "label": "CellBender cell size",
+          "description": "CellBender's estimate of the true number of RNA molecules in each cell after removing ambient contamination. Reflects actual cell RNA content rather than raw UMI counts.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_droplet_efficiency",
+          "label": "CellBender droplet efficiency",
+          "description": "CellBender's estimate of how efficiently each droplet captured RNA molecules. Higher values indicate more reliable RNA sampling within individual droplets.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/src/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json b/src/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
new file mode 100644
index 0000000..9dc7a8b
--- /dev/null
+++ b/src/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
@@ -0,0 +1,91 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "cell_area",
+          "label": "Segmented cell area",
+          "description": "Area of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "nucleus_ratio",
+          "label": "Nucleus Ratio",
+          "description": "Ratio of the nucleus area to the segmented cell area.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
\ No newline at end of file
diff --git a/src/ingestion_qc/h5mu_to_qc_json/script.py b/src/ingestion_qc/h5mu_to_qc_json/script.py
new file mode 100644
index 0000000..1706eb1
--- /dev/null
+++ b/src/ingestion_qc/h5mu_to_qc_json/script.py
@@ -0,0 +1,385 @@
+import json
+import pandas as pd
+from pathlib import Path
+import anndata as ad
+import h5py
+import sys
+import os
+import shutil
+
+## VIASH START
+# inputs = list(Path("data/sample_data/sample_data").glob("*.h5mu"))
+# output = "data/sample-data.json"
+inputs = list(Path("resources_test_after_running_script/qc_sample_data").glob("*.qc.h5mu"))
+output = "tmp.json"
+par = {
+    "input": sorted([str(x) for x in inputs]),
+    # "input": ["resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu", "resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu"],
+    "output": "sc_data.json",
+    "output_reporting_json": "sc_report_structure.json",
+    "modality": "rna",
+    "ingestion_method": "cellranger_multi",
+    "obs_sample_id": "sample_id",
+    "obs_total_counts": "total_counts",
+    "obs_num_nonzero_vars": "num_nonzero_vars",
+    "obs_fraction_mitochondrial": "fraction_mitochondrial",
+    "obs_fraction_ribosomal": "fraction_ribosomal",
+    "min_total_counts": 20,
+    "min_num_nonzero_vars": 20,
+    "obs_cellbender": [
+        "cellbender_background_fraction",
+        "cellbender_cell_probability",
+        "cellbender_cell_size",
+        "cellbender_droplet_efficiency",
+    ],
+    "uns_cellranger_metrics": "metrics_cellranger",
+    "obs_metadata": ["cell_type"],
+    "obs_nucleus_area": "nucleus_area",
+    "obs_cell_area": "cell_area",
+    "obs_x_coord": "x_coord",
+    "obs_y_coord": "y_coord",
+    "obs_control_probe_counts": "control_probe_counts",
+    "obs_control_codeword_counts": "control_codeword_counts"
+}
+meta = {
+    "resources_dir": os.path.abspath("src/ingestion_qc/h5mu_to_qc_json"),
+}
+i = 0
+mudata_file = par["input"][i]
+
+sys.path.append("src/utils")
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+par["obs_cellbender"] = {} if not par["obs_cellbender"] else par["obs_cellbender"]
+
+
+def transform_df(df):
+    """Transform a DataFrame into the annotation object format."""
+    columns = []
+    for name in df.columns:
+        data = df[name]
+
+        # Determine dtype
+        if pd.api.types.is_integer_dtype(data):
+            dtype = "integer"
+        elif pd.api.types.is_float_dtype(data):
+            dtype = "numeric"
+        elif pd.api.types.is_categorical_dtype(data):
+            dtype = "categorical"
+        else:
+            raise ValueError(f"Unknown/unsupported data type for column {name}")
+
+        column_info = {"name": name, "dtype": dtype}
+
+        if dtype == "categorical":
+            column_info["data"] = data.cat.codes.tolist()
+            column_info["categories"] = data.cat.categories.tolist()
+        else:
+            column_info["data"] = [None if pd.isna(x) else x for x in data]
+
+        columns.append(column_info)
+
+    return {"num_rows": len(df), "num_cols": len(df.columns), "columns": columns}
+
+
+def check_optional_obs_keys(obs, keys, message):
+    missing_keys = [key for key in keys if key not in obs.columns]
+    if missing_keys:
+        logger.info(f"Missing keys in obs: {', '.join(missing_keys)}. {message}")
+
+
+def transform_cellranger_metrics(uns, sample_id):
+    if not par["uns_cellranger_metrics"] in uns:
+        raise ValueError(f"Could not find cellranger metrics in uns: {par['uns_cellranger_metrics']}. Provide correct value for --uns_cellranger_metrics or make sure data was ingested using CellRanger multi.")
+
+    cellranger_metrics = (
+        uns[par["uns_cellranger_metrics"]]
+        .pivot_table(
+            index=[],
+            columns="Metric Name",
+            values="Metric Value",
+            aggfunc="first",
+        )
+        .reset_index(drop=True)
+    )
+
+    cellranger_metrics.columns.name = None
+    # Remove thousands separator and convert to numeric
+    cellranger_metrics = cellranger_metrics.map(
+        lambda x: (
+            pd.to_numeric(x.replace(",", ""), errors="coerce")
+            if isinstance(x, str)
+            else x
+        )
+    )
+    # Replace spaces with underscores in column names
+    cellranger_metrics.columns = cellranger_metrics.columns.str.replace(" ", "_")
+    for col in cellranger_metrics.columns:
+        cellranger_metrics[col] = pd.to_numeric(cellranger_metrics[col], errors="coerce")
+    cellranger_metrics["sample_id"] = [sample_id[0]]
+
+    return cellranger_metrics
+
+
+def format_cellbender_columns(mod_obs):
+    # Check if celbender was run on the dataset
+    if par["obs_cellbender"]:
+        check_optional_obs_keys(mod_obs, par["obs_cellbender"], "Run cellbender first to include these metrics.")
+
+    cellbender_obs_keys = [column for column in par["obs_cellbender"] if column in mod_obs]
+
+    for key in cellbender_obs_keys:
+        if not pd.api.types.is_float_dtype(mod_obs[key]):
+            try:
+                mod_obs[key] = mod_obs[key].astype("float16")
+            except ValueError:
+                raise ValueError(f"Could not convert column {key} to a float dtype. Please make sure all cellbender metrics are numeric.")
+
+    return cellbender_obs_keys, mod_obs
+
+
+def format_required_columns(required_keys, mod_obs):
+
+    for key in required_keys:
+        if not pd.api.types.is_numeric_dtype(mod_obs[key]):
+            raise ValueError(f"Column {key} must be a numeric dtype.")
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_total_counts"]]):
+        logger.info(f"Converting {par['obs_total_counts']} from {mod_obs[par['obs_total_counts']].dtype} to integer dtype...")
+        mod_obs[par["obs_total_counts"]] = mod_obs[par["obs_total_counts"]].astype(int)
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_num_nonzero_vars"]]):
+        logger.info(f"Converting {par['obs_num_nonzero_vars']} from {mod_obs[par['obs_num_nonzero_vars']].dtype} to integer dtype...")
+        mod_obs[par["obs_num_nonzero_vars"]] = mod_obs[par["obs_num_nonzero_vars"]].astype(int)
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_mitochondrial"]]):
+        logger.info(f"Converting {par['obs_fraction_mitochondrial']} from {mod_obs[par['obs_fraction_mitochondrial']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_mitochondrial"]] = mod_obs[par["obs_fraction_mitochondrial"]].astype("float16")
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_ribosomal"]]):
+        logger.info(f"Converting {par['obs_fraction_ribosomal']} from {mod_obs[par['obs_fraction_ribosomal']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_ribosomal"]] = mod_obs[par["obs_fraction_ribosomal"]].astype("float16")
+
+    return mod_obs
+
+
+def format_categorical_columns(mod_obs):
+    # Fetch all categorical columns for grouping if no columns are provided
+    if not par["obs_metadata"]:
+        metadata_obs_keys = mod_obs.select_dtypes(include=["object", "category"]).columns.tolist()
+        if par["obs_sample_id"] in metadata_obs_keys:
+            metadata_obs_keys.remove(par["obs_sample_id"])
+    else:
+        check_optional_obs_keys(mod_obs, par["obs_metadata"], "Make sure requested metadata colmuns are present in obs.")
+        metadata_obs_keys = [key for key in par["obs_metadata"] if key in mod_obs]
+
+    for key in metadata_obs_keys:
+        if not isinstance(key, pd.CategoricalDtype):
+            logger.info(f"{key} is not a categorical dtype. Converting {key} from {mod_obs[key].dtype} to categorical dtype...")
+            mod_obs[key] = mod_obs[key].astype(str).astype("category")
+
+    return metadata_obs_keys, mod_obs
+
+
+def generate_cellranger_stats(mod_obs, uns, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Fetch and format cellbender columns
+    cellbender_obs_keys, mod_obs = format_cellbender_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in cellbender_obs_keys},
+            **{key: mod_obs[key] for key in metadata_obs_keys},
+        }
+    )
+
+    cellranger_stats = transform_cellranger_metrics(uns, sample_id)
+
+    return cell_rna_stats, cellranger_stats
+
+
+def format_xenium_columns(mod_obs):
+
+    mod_obs["nucleus_ratio"] = mod_obs[par["obs_nucleus_area"]] / mod_obs[par["obs_cell_area"]]
+
+    xenium_formatted_columns = [par["obs_cell_area"], "nucleus_ratio", "x_coord", "y_coord"]
+    for key in xenium_formatted_columns:
+        mod_obs[key] = mod_obs[key].astype("float16")
+
+    return mod_obs, xenium_formatted_columns
+
+
+def generate_xenium_stats(mod_obs, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format xenium-specific columns
+    mod_obs, xenium_formatted_columns = format_xenium_columns(mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in xenium_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+
+    return cell_rna_stats
+
+
+def concatenate_dataframes(dfs):
+    '''Concatenates a list of dataframes into a single dataframe, preserving categorical columns.'''
+    df = pd.concat(dfs, ignore_index=True)
+
+    # Find categorical columns that became object columms
+    for col in df.columns:
+        if any(df[col].dtype.name == 'category' for df in dfs if col in df.columns):
+            # Get all categorical series for this column
+            cat_series = [df[col] for df in dfs if col in df.columns and df[col].dtype.name == 'category']
+            if cat_series:
+                # Union the categories and apply to result
+                unioned = pd.api.types.union_categoricals(cat_series)
+                df[col] = pd.Categorical(df[col], categories=unioned.categories)
+    return df
+
+
+def main(par):
+    cell_stats_dfs = []
+    sample_stats_dfs = []
+    metrics_cellranger_dfs = []
+
+    for i, mudata_file in enumerate(par["input"]):
+        logger.info(f"Processing {mudata_file}")
+
+        # read h5mu file
+        file = h5py.File(mudata_file, "r")
+
+        # read the necessary info
+        grp_mod = file["mod"][par["modality"]]
+        mod_obs = ad.experimental.read_elem(grp_mod["obs"])
+        mod_obsm = ad.experimental.read_elem(grp_mod["obsm"])
+        uns = ad.experimental.read_elem(file["uns"])
+
+        # close the h5mu file
+        file.close()
+
+        barcodes_original_count = mod_obs.shape[0]
+
+        # Add coordinates to obs before filtering
+        if par["ingestion_method"] == "xenium":
+            mod_obs["x_coord"] = mod_obsm["spatial"][:, 0]
+            mod_obs["y_coord"] = mod_obsm["spatial"][:, 1]
+
+        # Pre-filter cells
+        logger.info("Pre-filtering cells based on counts...")
+        if "min_total_counts" in par:
+            mod_obs = mod_obs[mod_obs["total_counts"] >= par["min_total_counts"]]
+        if "min_num_nonzero_vars" in par:
+            mod_obs = mod_obs[mod_obs["num_nonzero_vars"] >= par["min_num_nonzero_vars"]]
+        barcodes_filtered_count = mod_obs.shape[0]
+
+        # Detect sample id's
+        logger.info("Detecting sample id's...")
+        sample_id = (
+            mod_obs[par["obs_sample_id"]].tolist()
+            if par["obs_sample_id"] in mod_obs.columns
+            else [f"sample_{i}"] * mod_obs.shape[0]
+        )
+
+        # Generating sample summary statistics
+        logger.info("Generating sample summary statistics...")
+        required_keys = [
+            par["obs_total_counts"],
+            par["obs_num_nonzero_vars"],
+            par["obs_fraction_mitochondrial"],
+            par["obs_fraction_ribosomal"]
+            ]
+        missing_keys = [key for key in required_keys if key not in mod_obs.columns]
+        if missing_keys:
+            raise ValueError(f"Missing keys in obs: {', '.join(missing_keys)}")
+
+        sample_summary = {
+            "sample_id": pd.Categorical([sample_id[0]]),
+            "rna_num_barcodes": [barcodes_original_count],
+            "rna_num_barcodes_filtered": [barcodes_filtered_count],
+            "rna_sum_total_counts": [mod_obs[par["obs_total_counts"]].sum()],
+            "rna_median_total_counts": [mod_obs[par["obs_total_counts"]].median()],
+            "rna_overall_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].sum()],
+            "rna_median_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].median()],
+        }
+
+        if par["ingestion_method"] == "xenium":
+            sample_summary["control_probe_percentage"] = mod_obs[par["obs_control_probe_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+            sample_summary["negative_decoding_percentage"] = mod_obs[par["obs_control_codeword_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+
+        sample_summary_stats = pd.DataFrame(sample_summary)
+
+        if par["ingestion_method"] == "cellranger_multi":
+            cell_rna_stats, cellranger_stats = generate_cellranger_stats(mod_obs, uns, sample_id, required_keys)
+            metrics_cellranger_dfs.append(cellranger_stats)
+
+        if par["ingestion_method"] == "xenium":
+            cell_rna_stats = generate_xenium_stats(mod_obs, sample_id, required_keys)
+
+        cell_stats_dfs.append(cell_rna_stats)
+        sample_stats_dfs.append(sample_summary_stats)
+
+    # Combine dataframes of all samples
+    logger.info("Combining data of all samples into single object...")
+    combined_cell_stats = concatenate_dataframes(cell_stats_dfs)
+    combined_sample_stats = concatenate_dataframes(sample_stats_dfs)
+    if par["ingestion_method"] == "cellranger_multi":
+        combined_metrics_cellranger = concatenate_dataframes(metrics_cellranger_dfs)
+
+    report_categories = [combined_cell_stats, combined_sample_stats]
+
+    if par["ingestion_method"] == "cellranger_multi":
+        report_categories.append(combined_metrics_cellranger)
+
+    for df in report_categories:
+        df["sample_id"] = pd.Categorical(df["sample_id"])
+
+    output = {
+        "cell_rna_stats": transform_df(combined_cell_stats),
+        "sample_summary_stats": transform_df(combined_sample_stats)
+    }
+
+    if par["ingestion_method"] == "cellranger_multi":
+        output["metrics_cellranger_stats"] = transform_df(combined_metrics_cellranger)
+
+    logger.info(f"Writing output data json to {par['output']}")
+    output_path = Path(par["output"])
+    with open(output_path, "w") as f:
+        json.dump(output, f, indent=2)
+
+    report_structures = {
+        "cellranger_multi": os.path.join(meta["resources_dir"], "report_structure/cellranger.json"),
+        "xenium": os.path.join(meta["resources_dir"], "report_structure/xenium.json")
+    }
+
+    logger.info(f"Writing output report structure json to {par['output_reporting_json']}")
+    shutil.copy(report_structures[par["ingestion_method"]], par["output_reporting_json"])
+
+
+if __name__ == "__main__":
+    main(par)
diff --git a/src/ingestion_qc/h5mu_to_qc_json/test.py b/src/ingestion_qc/h5mu_to_qc_json/test.py
new file mode 100644
index 0000000..67d8e27
--- /dev/null
+++ b/src/ingestion_qc/h5mu_to_qc_json/test.py
@@ -0,0 +1,144 @@
+import pytest
+import os
+import json
+import sys
+import numpy as np
+
+## VIASH START
+meta = {
+    "resources_dir": "resources_test",
+    "executable": "./target/executable/ingestion_qc/h5mu_to_qc_json/h5mu_to_qc_json"
+}
+## VIASH END
+
+
+def test_cellranger_execution(run_component, tmp_path):
+    output_json_path = tmp_path / "output.json"
+    output_reporting_json_path = tmp_path / "output_reporting.json"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/resources_test/qc_sample_data/sample_one.qc.cellbender.h5mu",
+            "--input", meta["resources_dir"] + "/resources_test/qc_sample_data/sample_two.qc.cellbender.h5mu",
+            "--ingestion_method", "cellranger_multi",
+            "--output", output_json_path,
+            "--output_reporting_json", output_reporting_json_path
+        ]
+    )
+
+    assert os.path.exists(output_json_path), "Output file was not created"
+
+    with open(output_json_path, "r") as f:
+        output_json_dict = json.load(f)
+
+    assert output_json_dict.keys() == {"cell_rna_stats", "sample_summary_stats", "metrics_cellranger_stats"}
+
+    column_names_cell = [col["name"] for col in output_json_dict["cell_rna_stats"]["columns"]]
+    expected_column_names = [
+        "sample_id", "total_counts", "num_nonzero_vars",
+        "fraction_mitochondrial",  "fraction_ribosomal",
+        "cellbender_background_fraction", "cellbender_cell_probability",
+        "cellbender_cell_size", "cellbender_droplet_efficiency",
+        "donor_id", "cell_type", "batch", "condition"
+        ]
+
+    assert np.all([column in column_names_cell for column in expected_column_names])
+
+    for key in output_json_dict.keys():
+        assert output_json_dict[key].keys() == {"num_rows", "num_cols", "columns"}
+        for col in output_json_dict[key]["columns"]:
+            assert {"name", "dtype", "data"}.issubset(col.keys())
+
+
+def test_set_filters(run_component, tmp_path):
+    output_json_path = tmp_path / "output.json"
+    output_reporting_json_path = tmp_path / "output_reporting.json"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/resources_test/qc_sample_data/sample_one.qc.cellbender.h5mu",
+            "--input", meta["resources_dir"] + "/resources_test/qc_sample_data/sample_two.qc.cellbender.h5mu",
+            "--ingestion_method", "cellranger_multi",
+            "--output", output_json_path,
+            "--output_reporting_json", output_reporting_json_path,
+            "--obs_sample_id", "sample_id",
+            "--obs_total_counts", "total_counts",
+            "--obs_num_nonzero_vars", "num_nonzero_vars",
+            "--obs_fraction_mitochondrial", "fraction_mitochondrial",
+            "--obs_fraction_ribosomal", "fraction_ribosomal",
+            "--min_total_counts", "20",
+            "--min_num_nonzero_vars", "20",
+            "--obs_metadata", "cell_type"
+        ]
+    )
+
+    assert os.path.exists(output_json_path), "Output file was not created"
+
+    with open(output_json_path, "r") as f:
+        output_json_dict = json.load(f)
+
+    assert output_json_dict.keys() == {"cell_rna_stats", "sample_summary_stats", "metrics_cellranger_stats"}
+
+    column_names = [col["name"] for col in output_json_dict["cell_rna_stats"]["columns"]]
+    expected_column_names = [
+        "sample_id", "total_counts", "num_nonzero_vars",
+        "fraction_mitochondrial",  "fraction_ribosomal",
+        "cellbender_background_fraction", "cellbender_cell_probability",
+        "cellbender_cell_size", "cellbender_droplet_efficiency",
+        "cell_type"
+        ]
+    unexpected_column_names = ["batch", "condition", "donor_id"]
+    assert np.all([column in column_names for column in expected_column_names])
+    assert np.all([column not in column_names for column in unexpected_column_names])
+    for key in output_json_dict.keys():
+        assert output_json_dict[key].keys() == {"num_rows", "num_cols", "columns"}
+        for col in output_json_dict[key]["columns"]:
+            assert {"name", "dtype", "data"}.issubset(col.keys())
+
+    total_counts = next(col for col in output_json_dict["cell_rna_stats"]["columns"] if col["name"] == "total_counts")
+    assert min(total_counts["data"]) >= 20
+
+    num_nonzero_vars = next(col for col in output_json_dict["cell_rna_stats"]["columns"] if col["name"] == "num_nonzero_vars")
+    assert min(num_nonzero_vars["data"]) >= 20
+
+
+def test_xenium_execution(run_component, tmp_path):
+    output_json_path = tmp_path / "output.json"
+    output_reporting_json_path = tmp_path / "output_reporting.json"
+
+    run_component(
+        [
+            "--input", meta["resources_dir"] + "/resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu",
+            "--input", meta["resources_dir"] + "/resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu",
+            "--ingestion_method", "xenium",
+            "--min_num_nonzero_vars", "1",
+            "--output", output_json_path,
+            "--output_reporting_json", output_reporting_json_path
+        ]
+    )
+
+    assert os.path.exists(output_json_path), "Output file was not created"
+
+    with open(output_json_path, "r") as f:
+        output_json_dict = json.load(f)
+
+    assert output_json_dict.keys() == {"cell_rna_stats", "sample_summary_stats"}
+    assert "metrics_cellranger_stats" not in output_json_dict.keys()
+
+    column_names_cell = [col["name"] for col in output_json_dict["cell_rna_stats"]["columns"]]
+    expected_column_names = [
+        "sample_id", "total_counts", "num_nonzero_vars",
+        "fraction_mitochondrial",  "fraction_ribosomal",
+        "cell_area", "nucleus_ratio",
+        "x_coord", "y_coord", "cell_id", "segmentation_method", "region"
+        ]
+    assert np.all([column in column_names_cell for column in expected_column_names])
+
+    for key in output_json_dict.keys():
+        assert output_json_dict[key].keys() == {"num_rows", "num_cols", "columns"}
+        for col in output_json_dict[key]["columns"]:
+            assert {"name", "dtype", "data"}.issubset(col.keys())
+
+
+if __name__ == "__main__":
+    sys.exit(pytest.main([__file__]))
diff --git a/src/utils/setup_logger.py b/src/utils/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/src/utils/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/src/workflows/generate_qc_report/config.vsh.yaml b/src/workflows/generate_qc_report/config.vsh.yaml
new file mode 100644
index 0000000..a306f0b
--- /dev/null
+++ b/src/workflows/generate_qc_report/config.vsh.yaml
@@ -0,0 +1,135 @@
+name: generate_qc_report
+namespace: workflows
+description: Run the ingestion QC report generation
+authors: 
+  - __merge__: /src/authors/jakub_majercik.yaml
+    roles: [author]
+  - __merge__: /src/authors/dorien_roosen.yaml
+    roles: [author]
+  - __merge__: /src/authors/robrecht_cannoodt.yaml
+    roles: [author]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [contributor]
+argument_groups:
+  # TO DO: it would be nice if the sample metadata was already
+  # included in the h5mu files, so that we don't need to pass it.
+  - name: Inputs
+    arguments:
+      - name: --id
+        type: string
+        required: false
+        direction: input
+        description: |
+          The sample IDs to include in the report. If not provided,
+          the sample IDs will be extracted from the h5mu files.
+        example: sample1
+      - name: --input
+        type: file
+        required: true
+        direction: input
+        description: The input h5mu files.
+        example: path/to/file1.h5mu
+      - name: --ingestion_method
+        type: string
+        required: true
+        choices:
+          - cellranger_multi
+          - xenium
+      - name: --sample_metadata
+        type: file
+        required: false
+        direction: input
+        description: |
+          The sample metadata file corresponding to .obs fields in the h5mu input files, to be used for grouping in the report.
+        example: path/to/file.csv
+      - name: --max_samples_per_report
+        type: integer
+        default: 20
+        description: |
+          The maximum number of samples to be included per report. 
+          Multiple reports will be generated (with samples equally divided over all reports) if number of input samples exceeds this threshold.
+  - name: Options
+    arguments:
+      - name: "--var_gene_names"
+        example: "gene_symbol"
+        type: string
+        description: |
+          The column name in the .var h5mu files that contains the gene names. If not provided, .var_names will be used.
+      - name: --obs_metadata
+        type: string
+        multiple: true
+        description: The metadata keys in the h5mu .obs to include in the report.
+        example: [donor_id, cell_type, batch, condition]
+
+  - name: QC options
+    arguments:
+      - name: "--var_name_mitochondrial_genes"
+        type: string
+        required: false
+        default: "mitochondrial"
+        description: |
+          In which .var slot to store a boolean array corresponding the mitochondrial genes.
+      - name: "--var_name_ribosomal_genes"
+        type: string
+        required: false
+        default: "ribosomal"
+        description: |
+          In which .var slot to store a boolean array corresponding the ribosomal genes.
+  
+  - name: Cellbender options
+    arguments:
+      - name: "--run_cellbender"
+        type: boolean
+        required: false
+        description: Whether to run cellbender or not.
+        default: false
+      - name: "--cellbender_epochs"
+        type: integer
+        required: false
+        description: Number of epochs to train cellbender. 
+        default: 150
+  
+  - name: Outputs
+    arguments:
+      - name: --output_qc_report
+        type: file
+        required: true
+        multiple: true
+        direction: output
+        description: The output HTML report
+        example: path/to/file.html
+      - name: --output_processed_h5mu
+        type: file
+        required: true
+        direction: output
+        description: Folder containing the processed h5mu files.
+        default: qc_h5mu
+
+resources:
+  - type: nextflow_script
+    entrypoint: run_wf
+    path: main.nf
+
+test_resources:
+  - type: nextflow_script
+    path: test.nf
+    entrypoint: test_no_cellbender
+  - type: nextflow_script
+    path: test.nf
+    entrypoint: test_with_cellbender
+
+dependencies:
+  - name: metadata/add_id
+    repository: openpipeline
+  - name: workflows/qc/qc
+    repository: openpipeline
+  - name: correction/cellbender_remove_background
+    alias: cellbender
+    repository: openpipeline
+  - name: ingestion_qc/h5mu_to_qc_json
+  - name: ingestion_qc/generate_html
+  - name: move_files_to_directory
+    repository: craftbox
+
+runners:
+  - type: nextflow
diff --git a/src/workflows/generate_qc_report/integration_test.sh b/src/workflows/generate_qc_report/integration_test.sh
new file mode 100755
index 0000000..ee15473
--- /dev/null
+++ b/src/workflows/generate_qc_report/integration_test.sh
@@ -0,0 +1,37 @@
+#!/usr/bin/env bash
+
+# get the root of the directory
+REPO_ROOT=$(git rev-parse --show-toplevel)
+
+# ensure that the command below is run from the root of the repository
+cd "$REPO_ROOT"
+
+viash ns build --setup cb -q generate_qc_report
+
+nextflow run . \
+  -main-script src/workflows/generate_qc_report/test.nf \
+  -profile docker,no_publish,local \
+  -entry test_no_cellbender \
+  -c src/configs/labels_ci.config \
+  -resume
+
+nextflow run . \
+  -main-script src/workflows/generate_qc_report/test.nf \
+  -profile docker,no_publish,local \
+  -entry test_xenium \
+  -c src/configs/labels_ci.config \
+  -resume
+
+nextflow run . \
+  -main-script src/workflows/generate_qc_report/test.nf \
+  -profile docker,no_publish,local \
+  -entry test_with_cellbender \
+  -c src/configs/labels_ci.config \
+  -resume
+
+nextflow run . \
+  -main-script src/workflows/generate_qc_report/test.nf \
+  -profile docker,no_publish,local \
+  -entry test_multiple_reports \
+  -c src/configs/labels_ci.config \
+  -resume
diff --git a/src/workflows/generate_qc_report/main.nf b/src/workflows/generate_qc_report/main.nf
new file mode 100644
index 0000000..f1dd5c2
--- /dev/null
+++ b/src/workflows/generate_qc_report/main.nf
@@ -0,0 +1,229 @@
+workflow run_wf {
+  take: input_ch
+  main:
+  qc_ch = input_ch
+    // store join id
+    | map { id, state ->
+      [id, state + [_meta: [join_id: id]]]
+    }
+
+    // add sample ids to each state
+    | add_id.run(
+      fromState: [
+        input_id: "id", 
+        input: "input"
+      ],
+      args: [
+        obs_output: "sample_id"
+      ],
+      toState: [ "input": "output" ]
+    )
+
+    // run cellbender
+    | cellbender.run(
+      runIf: {id, state -> state.run_cellbender},
+      fromState: [
+        id: "id",
+        input: "input",
+        epochs: "cellbender_epochs",
+      ],
+      args: [
+        obs_background_fraction: "cellbender_background_fraction",
+        obs_cell_probability: "cellbender_cell_probability",
+        obs_droplet_efficiency: "cellbender_droplet_efficiency",
+        obs_cell_size: "cellbender_cell_size",
+      ],
+      toState: ["input": "output"]
+    )
+
+    // run qc on each sample
+    | qc.run(
+      fromState: [
+        id: "id",
+        input: "input",
+        var_gene_names: "var_gene_names"
+      ],
+      args: [
+        var_name_mitochondrial_genes: "mitochondrial",
+        var_name_ribosomal_genes: "ribosomal",
+        output_obs_num_nonzero_vars: "num_nonzero_vars",
+        output_obs_total_counts_vars: "total_counts"
+      ],
+      toState: { id, output, state ->
+        def keysToRemove = ["var_gene_names", "var_name_mitochondrial_genes", "var_name_ribosomal_genes", "run_cellbender", "cellbender_epochs"]
+        def newState = state.findAll{it.key !in keysToRemove}
+        newState + ["input": output.output]
+      }
+    )
+
+    | joinStates { ids, states ->
+      def newId = "qc_data"
+      // gather keys with unique values across states that should be combined
+      def new_state_non_unique_values = [
+        input: states.collect{it.input},
+        join_ids: states.collect{it._meta.join_id},
+        _meta: [join_id: ids[0]]
+      ]
+      // gather keys from different states
+      def all_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+        }.minus(["output", "id", "input", "_meta"])
+      // Create the new state from the keys, values should be the same across samples
+      def new_state = all_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across samples. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            // take the unique value from the set (there is only one)
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+        }
+      def data_state = new_state_non_unique_values + new_state
+      [ newId, data_state ]
+    }
+
+  processed_files_ch = qc_ch
+
+    // move all processed h5mu files to the same folder
+    | move_files_to_directory.run(
+      fromState: [
+        input: "input",
+        output: "output_processed_h5mu"
+      ],
+      toState: [ "output_processed_h5mu": "output" ]
+    )
+    | setState(["output_processed_h5mu"])
+
+  report_ch = qc_ch
+    // group the processed samples to generate one or multiple reports
+    | flatMap { id, state ->
+
+        // calculate number of reports to be generated and number of samples per report
+        def totalInputs = state.input.size()
+        def maxSamplesPerGroup = state.max_samples_per_report
+        def numGroups = Math.max(1, Math.ceil(totalInputs / maxSamplesPerGroup) as Integer)
+        def baseSamplesPerGroup = totalInputs.intdiv(numGroups)
+        def remainder = totalInputs % numGroups
+
+        println "Dividing ${totalInputs} sample(s) over ${numGroups} report(s) (max ${maxSamplesPerGroup} per report)"
+        
+        // sort inputs to make grouping deterministic
+        def inputs = []
+        for (int i = 0; i < state.input.size(); i++) {
+            inputs << [input: state.input[i], _meta: [join_id: state.join_ids[i]]]
+        }
+
+        def sortedInputs = inputs.sort { it._meta.join_id }
+  
+        def groups = []
+        def itemIndex = 0
+
+        // create one channel per report
+        (0..<numGroups).each { groupNum ->
+            def samplesInGroup = baseSamplesPerGroup + (groupNum < remainder ? 1 : 0)
+            def groupItems = sortedInputs[itemIndex..<(itemIndex + samplesInGroup)]
+            
+            def newId = "combined_${groupNum + 1}_of_${numGroups}"
+            def newState = state.clone()  // Copy all the original state
+            
+            // Override the input and _meta with the grouped items
+            newState.input = groupItems.collect { it.input }
+            newState._meta = groupItems[0]._meta
+            
+            println "Group ${groupNum + 1}: ${samplesInGroup} samples - ${newState._meta}"
+            
+            groups << [newId, newState]
+            itemIndex += samplesInGroup
+        }
+        
+        return groups
+    }
+
+    // Set aside output for QC report instructions
+    | map { id, state -> 
+      def new_state = state + ["output_reporting_json": "reporting_json.json"]
+      [id, new_state]
+    }
+
+    // Set report filter settings
+    | map { id, state -> 
+      def conditionalValues = [
+        "cellranger_multi": [
+          "min_total_counts": 10,
+          "min_num_nonzero_vars": 10
+        ],
+        "xenium": [
+          "min_total_counts": 10, 
+          "min_num_nonzero_vars": 1
+        ]
+      ]
+      
+      def method = state.ingestion_method
+      def additionalParams = conditionalValues[method]
+      
+      [ id, state + additionalParams ]
+    }
+
+    | view {"After setting filters: $it"}
+
+    // generate qc json
+    | h5mu_to_qc_json.run(
+      fromState: [
+        input: "input",
+        ingestion_method: "ingestion_method",
+        obs_metadata: "obs_metadata",
+        min_total_counts: "min_total_counts",
+        min_num_nonzero_vars: "min_num_nonzero_vars"
+      ],
+      args: [
+        obs_sample_id: "sample_id",
+        obs_total_counts: "total_counts",
+        obs_num_nonzero_vars: "num_nonzero_vars",
+        obs_fraction_mitochondrial: "fraction_mitochondrial",
+        obs_fraction_ribosomal: "fraction_ribosomal",
+      ],
+      toState: [
+        output: "output",
+        output_reporting_json: "output_reporting_json"
+      ]
+    )
+
+    // generate html report
+    | generate_html.run(
+      fromState: [ 
+        input_data: "output",
+        input_structure: "output_reporting_json"
+      ],
+      toState: [
+        output_qc_report: "output_qc_report"
+      ]
+    )
+
+    // collect the reports into a single channel
+    | joinStates { ids, states ->
+      def newId = "qc_report"
+      def report_state = [
+        output_qc_report: states.collect{it.output_qc_report},
+        _meta: states[0]._meta
+      ]
+      [ newId, report_state ]
+    }
+    
+  output_ch = report_ch.mix(processed_files_ch)
+
+    | joinStates { ids, states ->
+
+      assert states.size() == 2, "Expected 2 states, but got ${states.size()}"
+      assert ids.contains('qc_report'), "Expected one channel to have the id `qc_report`, but got ${ids}"
+      assert ids.contains('qc_data'), "Expected one channel to have the id `qc_data`, but got ${ids}"
+
+      def newId = "combined"
+      def combined_state = states[0] + states [1]
+
+      [ newId, combined_state ]
+    }
+
+  emit: output_ch
+}
\ No newline at end of file
diff --git a/src/workflows/generate_qc_report/nextflow.config b/src/workflows/generate_qc_report/nextflow.config
new file mode 100644
index 0000000..b0dc0d8
--- /dev/null
+++ b/src/workflows/generate_qc_report/nextflow.config
@@ -0,0 +1,10 @@
+manifest {
+  nextflowVersion = '!>=20.12.1-edge'
+}
+
+params {
+  rootDir = java.nio.file.Paths.get("$projectDir/../../../").toAbsolutePath().normalize().toString()
+}
+
+// include common settings
+includeConfig("${params.rootDir}/src/configs/labels.config")
\ No newline at end of file
diff --git a/src/workflows/generate_qc_report/test.nf b/src/workflows/generate_qc_report/test.nf
new file mode 100644
index 0000000..4d4090a
--- /dev/null
+++ b/src/workflows/generate_qc_report/test.nf
@@ -0,0 +1,222 @@
+nextflow.enable.dsl=2
+targetDir = params.rootDir + "/target/nextflow/workflows"
+
+include { generate_qc_report } from targetDir + "/generate_qc_report/main.nf"
+
+params.resources_test = "s3://openpipelines-bio/openpipeline_incubator/resources_test/"
+
+workflow test_no_cellbender {
+
+  resources_test_file = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_1",
+        input: resources_test_file.resolve("qc_sample_data/sample_one.qc.h5mu"),
+        run_cellbender: false,
+        ingestion_method: "cellranger_multi",
+        var_gene_names: "gene_symbol",
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_2",
+        input: resources_test_file.resolve("qc_sample_data/sample_two.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        var_gene_names: "gene_symbol",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ]
+    ])
+
+    | map{ state -> [state.id, state] }
+    | generate_qc_report
+
+    | view { output ->
+        assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+        def id = output[0]
+        def state = output [1]
+        assert id == "combined": "Output ID should be `combined`"
+        assert state instanceof Map : "State should be a map. Found: ${state}"
+        assert state.containsKey("output_qc_report"): "Output should contain key `output_qc_report`"
+        assert state.containsKey("output_processed_h5mu"): "Output should contain key `output_processed_h5mu`"
+        assert state.output_qc_report.size() == 1 : "Expected exactly one output HTML file to be generated"
+        assert state.output_qc_report.every { it.isFile()} : "All output HTML report file should exist"
+        assert state.output_processed_h5mu.isDirectory() : "Output directory should exist"
+        def files = state.output_processed_h5mu.listFiles().findAll { it.isFile() }
+        assert files.size() == 2 : "Output directory should contain exactly 2 files, but found ${files.size()} files"
+        "Output: $output"
+    }
+}
+
+workflow test_xenium {
+
+  resources_test_file = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_one",
+        input: resources_test_file.resolve("spatial_qc_sample_data/xenium_tiny.qc.h5mu"),
+        run_cellbender: false,
+        ingestion_method: "xenium",
+        var_gene_names: "gene_ids",
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_two",
+        input: resources_test_file.resolve("spatial_qc_sample_data/xenium_tiny.qc.h5mu"),
+        ingestion_method: "xenium",
+        var_gene_names: "gene_ids",
+        run_cellbender: false,
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ]
+    ])
+
+    | map{ state -> [state.id, state] }
+    | generate_qc_report
+
+    | view { output ->
+        assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+        def id = output[0]
+        def state = output [1]
+        assert id == "combined": "Output ID should be `combined`"
+        assert state instanceof Map : "State should be a map. Found: ${state}"
+        assert state.containsKey("output_qc_report"): "Output should contain key `output_qc_report`"
+        assert state.containsKey("output_processed_h5mu"): "Output should contain key `output_processed_h5mu`"
+        assert state.output_qc_report.size() == 1 : "Expected exactly one output HTML file to be generated"
+        assert state.output_qc_report.every { it.isFile()} : "All output HTML report file should exist"
+        assert state.output_processed_h5mu.isDirectory() : "Output directory should exist"
+        def files = state.output_processed_h5mu.listFiles().findAll { it.isFile() }
+        assert files.size() == 2 : "Output directory should contain exactly 2 files, but found ${files.size()} files"
+        "Output: $output"
+    }
+}
+
+workflow test_with_cellbender {
+
+  resources_test_file = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_one",
+        input: resources_test_file.resolve("qc_sample_data/sample_one.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        var_gene_names: "gene_symbol",
+        run_cellbender: true,
+        cellbender_epochs: 1,
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_two",
+        input: resources_test_file.resolve("qc_sample_data/sample_two.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        var_gene_names: "gene_symbol",
+        run_cellbender: true,
+        cellbender_epochs: 1,
+        output_html: "report.html",
+        publish_dir: "test_out"
+      ]
+    ])
+
+    | map{ state -> [state.id, state] }
+    | generate_qc_report
+
+    | view { output ->
+        assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+        def id = output[0]
+        def state = output [1]
+        assert id == "combined": "Output ID should be `combined`"
+        assert state instanceof Map : "State should be a map. Found: ${state}"
+        assert state.containsKey("output_qc_report"): "Output should contain key `output_qc_report`"
+        assert state.containsKey("output_processed_h5mu"): "Output should contain key `output_processed_h5mu`"
+        assert state.output_qc_report.size() == 1 : "Expected exactly one output HTML file to be generated"
+        assert state.output_qc_report.every { it.isFile()} : "All output HTML report file should exist"
+        assert state.output_processed_h5mu.isDirectory() : "Output directory should exist"
+        def files = state.output_processed_h5mu.listFiles().findAll { it.isFile() }
+        assert files.size() == 2 : "Output directory should contain exactly 2 files, but found ${files.size()} files"
+        "Output: $output"
+    }
+}
+
+workflow test_multiple_reports {
+
+  resources_test_file = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_1",
+        input: resources_test_file.resolve("qc_sample_data/sample_one.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        max_samples_per_report: 2,
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_2",
+        input: resources_test_file.resolve("qc_sample_data/sample_two.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        max_samples_per_report: 2,
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_3",
+        input: resources_test_file.resolve("qc_sample_data/sample_one.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        max_samples_per_report: 2,
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_4",
+        input: resources_test_file.resolve("qc_sample_data/sample_two.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        max_samples_per_report: 2,
+        publish_dir: "test_out"
+      ],
+      [
+        id: "sample_5",
+        input: resources_test_file.resolve("qc_sample_data/sample_one.qc.h5mu"),
+        ingestion_method: "cellranger_multi",
+        run_cellbender: false,
+        metadata_obs_keys: ["donor_id", "cell_type", "batch", "condition"],
+        output_html: "report.html",
+        max_samples_per_report: 2,
+        publish_dir: "test_out"
+      ]
+    ])
+
+    | map{ state -> [state.id, state] }
+    | generate_qc_report
+
+    | view { output ->
+        assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+        def id = output[0]
+        def state = output [1]
+        assert id == "combined": "Output ID should be `combined`"
+        assert state instanceof Map : "State should be a map. Found: ${state}"
+        assert state.containsKey("output_qc_report"): "Output should contain key `output_qc_report`"
+        assert state.containsKey("output_processed_h5mu"): "Output should contain key `output_processed_h5mu`"
+        assert state.output_qc_report.size() == 3 : "Expected exactly one output HTML file to be generated"
+        assert state.output_qc_report.every { it.isFile()} : "All output HTML report file should exist"
+        assert state.output_processed_h5mu.isDirectory() : "Output directory should exist"
+        def files = state.output_processed_h5mu.listFiles().findAll { it.isFile() }
+        assert files.size() == 5 : "Output directory should contain exactly 5 files, but found ${files.size()} files"
+        "Output: $output"
+    }
+}
\ No newline at end of file
diff --git a/target/.build.yaml b/target/.build.yaml
new file mode 100644
index 0000000..e69de29
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
new file mode 100644
index 0000000..c88614a
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
@@ -0,0 +1,483 @@
+name: "h5mu_to_qc_json"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu file(s)"
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "The modality to use"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ingestion_method"
+    description: "Method that was used to ingest the data - this will define the structure\
+      \ of the report that is generated."
+    info: null
+    required: true
+    choices:
+    - "cellranger_multi"
+    - "xenium"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_sample_id"
+    description: "The key in the h5mu file that contains the sample ID. If not provided,\
+      \ each H5MU file will be considered as a separate sample."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_total_counts"
+    description: "The key in the h5mu .obs field that contains the total counts."
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_num_nonzero_vars"
+    description: "The key in the h5mu .obs field that contains the number of nonzero\
+      \ vars."
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_mitochondrial"
+    description: "The key in the h5mu .obs field that contains the fraction mitochondrial\
+      \ genes."
+    info: null
+    default:
+    - "fraction_mitochondrial"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_ribosomal"
+    description: "The key in the h5mu .obs field that contains the fraction ribosomal\
+      \ genes."
+    info: null
+    default:
+    - "fraction_ribosomal"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The output JSON file"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_reporting_json"
+    description: "The output JSON file that defines the QC report"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering & grouping options"
+  arguments:
+  - type: "integer"
+    name: "--min_total_counts"
+    description: "Minimum total counts for a cell to be included in the output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_num_nonzero_vars"
+    description: "Minimum number of nonzero vars for a cell to be included in the\
+      \ output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_metadata"
+    description: "The metadata keys in the h5mu .obs to include in the output JSON."
+    info: null
+    example:
+    - "donor_id;cell_type;batch;condition"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Options for CellRanger reports"
+  arguments:
+  - type: "string"
+    name: "--obs_cellbender"
+    description: "The cellbender keys in the h5mu .obs to include in the output JSON"
+    info: null
+    default:
+    - "cellbender_background_fraction"
+    - "cellbender_cell_probability"
+    - "cellbender_cell_size"
+    - "cellbender_droplet_efficiency"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_cellranger_metrics"
+    description: "The key in the h5mu file .uns that contains the cellranger metrics"
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options for Xenium reports"
+  arguments:
+  - type: "string"
+    name: "--obs_nucleus_area"
+    description: "The key in the h5mu .obs field that contains the nucleus area."
+    info: null
+    default:
+    - "nucleus_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_area"
+    description: "The key in the h5mu .obs field that contains the cell area."
+    info: null
+    default:
+    - "cell_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_x_coord"
+    description: "The key in the h5mu .obs field that contains the x coordinate."
+    info: null
+    default:
+    - "x_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_y_coord"
+    description: "The key in the h5mu .obs field that contains the y coordinate."
+    info: null
+    default:
+    - "y_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_probe_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ probes."
+    info: null
+    default:
+    - "control_probe_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_codeword_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ codewords."
+    info: null
+    default:
+    - "control_codeword_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "report_structure"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx\
+  \ and processed by the QC workflow, and generates:\n- A JSON file that contains\
+  \ the combined data for the QC report\n- A JSON file that defines the layout and\
+  \ structure of the QC report\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "resources_test"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml"
+  runner: "executable"
+  engine: "docker|native"
+  output: "target/_private/executable/ingestion_qc/h5mu_to_qc_json"
+  executable: "target/_private/executable/ingestion_qc/h5mu_to_qc_json/h5mu_to_qc_json"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/h5mu_to_qc_json b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/h5mu_to_qc_json
new file mode 100755
index 0000000..65b947e
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/h5mu_to_qc_json
@@ -0,0 +1,1939 @@
+#!/usr/bin/env bash
+
+# h5mu_to_qc_json build_main
+# 
+# This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+# work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+# Intuitive.
+# 
+# The component may contain files which fall under a different license. The
+# authors of this component should specify the license in the header of such
+# files, or include a separate license file detailing the licenses of all included
+# files.
+# 
+# Component authors:
+#  * Jakub Majercik (author)
+#  * Dorien Roosen (author)
+#  * Robrecht Cannoodt (author)
+#  * Weiwei Schultz (contributor)
+
+set -e
+
+if [ -z "$VIASH_TEMP" ]; then
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMP}
+  VIASH_TEMP=${VIASH_TEMP:-/tmp}
+fi
+
+# define helper functions
+# ViashQuote: put quotes around non flag values
+# $1     : unquoted string
+# return : possibly quoted string
+# examples:
+#   ViashQuote --foo      # returns --foo
+#   ViashQuote bar        # returns 'bar'
+#   Viashquote --foo=bar  # returns --foo='bar'
+function ViashQuote {
+  if [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+=.+$ ]]; then
+    echo "$1" | sed "s#=\(.*\)#='\1'#"
+  elif [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+$ ]]; then
+    echo "$1"
+  else
+    echo "'$1'"
+  fi
+}
+# ViashRemoveFlags: Remove leading flag
+# $1     : string with a possible leading flag
+# return : string without possible leading flag
+# examples:
+#   ViashRemoveFlags --foo=bar  # returns bar
+function ViashRemoveFlags {
+  echo "$1" | sed 's/^--*[a-zA-Z0-9_\-]*=//'
+}
+# ViashSourceDir: return the path of a bash file, following symlinks
+# usage   : ViashSourceDir ${BASH_SOURCE[0]}
+# $1      : Should always be set to ${BASH_SOURCE[0]}
+# returns : The absolute path of the bash file
+function ViashSourceDir {
+  local source="$1"
+  while [ -h "$source" ]; do
+    local dir="$( cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd )"
+    source="$(readlink "$source")"
+    [[ $source != /* ]] && source="$dir/$source"
+  done
+  cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd
+}
+# ViashFindTargetDir: return the path of the '.build.yaml' file, following symlinks
+# usage   : ViashFindTargetDir 'ScriptPath'
+# $1      : The location from where to start the upward search
+# returns : The absolute path of the '.build.yaml' file
+function ViashFindTargetDir {
+  local source="$1"
+  while [[ "$source" != "" && ! -e "$source/.build.yaml" ]]; do
+    source=${source%/*}
+  done
+  echo $source
+}
+# see https://en.wikipedia.org/wiki/Syslog#Severity_level
+VIASH_LOGCODE_EMERGENCY=0
+VIASH_LOGCODE_ALERT=1
+VIASH_LOGCODE_CRITICAL=2
+VIASH_LOGCODE_ERROR=3
+VIASH_LOGCODE_WARNING=4
+VIASH_LOGCODE_NOTICE=5
+VIASH_LOGCODE_INFO=6
+VIASH_LOGCODE_DEBUG=7
+VIASH_VERBOSITY=$VIASH_LOGCODE_NOTICE
+
+# ViashLog: Log events depending on the verbosity level
+# usage: ViashLog 1 alert Oh no something went wrong!
+# $1: required verbosity level
+# $2: display tag
+# $3+: messages to display
+# stdout: Your input, prepended by '[$2] '.
+function ViashLog {
+  local required_level="$1"
+  local display_tag="$2"
+  shift 2
+  if [ $VIASH_VERBOSITY -ge $required_level ]; then
+    >&2 echo "[$display_tag]" "$@"
+  fi
+}
+
+# ViashEmergency: log events when the system is unstable
+# usage: ViashEmergency Oh no something went wrong.
+# stdout: Your input, prepended by '[emergency] '.
+function ViashEmergency {
+  ViashLog $VIASH_LOGCODE_EMERGENCY emergency "$@"
+}
+
+# ViashAlert: log events when actions must be taken immediately (e.g. corrupted system database)
+# usage: ViashAlert Oh no something went wrong.
+# stdout: Your input, prepended by '[alert] '.
+function ViashAlert {
+  ViashLog $VIASH_LOGCODE_ALERT alert "$@"
+}
+
+# ViashCritical: log events when a critical condition occurs
+# usage: ViashCritical Oh no something went wrong.
+# stdout: Your input, prepended by '[critical] '.
+function ViashCritical {
+  ViashLog $VIASH_LOGCODE_CRITICAL critical "$@"
+}
+
+# ViashError: log events when an error condition occurs
+# usage: ViashError Oh no something went wrong.
+# stdout: Your input, prepended by '[error] '.
+function ViashError {
+  ViashLog $VIASH_LOGCODE_ERROR error "$@"
+}
+
+# ViashWarning: log potentially abnormal events
+# usage: ViashWarning Something may have gone wrong.
+# stdout: Your input, prepended by '[warning] '.
+function ViashWarning {
+  ViashLog $VIASH_LOGCODE_WARNING warning "$@"
+}
+
+# ViashNotice: log significant but normal events
+# usage: ViashNotice This just happened.
+# stdout: Your input, prepended by '[notice] '.
+function ViashNotice {
+  ViashLog $VIASH_LOGCODE_NOTICE notice "$@"
+}
+
+# ViashInfo: log normal events
+# usage: ViashInfo This just happened.
+# stdout: Your input, prepended by '[info] '.
+function ViashInfo {
+  ViashLog $VIASH_LOGCODE_INFO info "$@"
+}
+
+# ViashDebug: log all events, for debugging purposes
+# usage: ViashDebug This just happened.
+# stdout: Your input, prepended by '[debug] '.
+function ViashDebug {
+  ViashLog $VIASH_LOGCODE_DEBUG debug "$@"
+}
+
+# find source folder of this component
+VIASH_META_RESOURCES_DIR=`ViashSourceDir ${BASH_SOURCE[0]}`
+
+# find the root of the built components & dependencies
+VIASH_TARGET_DIR=`ViashFindTargetDir $VIASH_META_RESOURCES_DIR`
+
+# define meta fields
+VIASH_META_NAME="h5mu_to_qc_json"
+VIASH_META_FUNCTIONALITY_NAME="h5mu_to_qc_json"
+VIASH_META_EXECUTABLE="$VIASH_META_RESOURCES_DIR/$VIASH_META_NAME"
+VIASH_META_CONFIG="$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+VIASH_META_TEMP_DIR="$VIASH_TEMP"
+
+
+
+# initialise variables
+VIASH_MODE='run'
+VIASH_ENGINE_ID='docker'
+
+######## Helper functions for setting up Docker images for viash ########
+# expects: ViashDockerBuild
+
+# ViashDockerInstallationCheck: check whether Docker is installed correctly
+#
+# examples:
+#   ViashDockerInstallationCheck
+function ViashDockerInstallationCheck {
+  ViashDebug "Checking whether Docker is installed"
+  if [ ! command -v docker &> /dev/null ]; then
+    ViashCritical "Docker doesn't seem to be installed. See 'https://docs.docker.com/get-docker/' for instructions."
+    exit 1
+  fi
+
+  ViashDebug "Checking whether the Docker daemon is running"
+  local save=$-; set +e
+  local docker_version=$(docker version --format '{{.Client.APIVersion}}' 2> /dev/null)
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashCritical "Docker daemon does not seem to be running. Try one of the following:"
+    ViashCritical "- Try running 'dockerd' in the command line"
+    ViashCritical "- See https://docs.docker.com/config/daemon/"
+    exit 1
+  fi
+}
+
+# ViashDockerRemoteTagCheck: check whether a Docker image is available 
+# on a remote. Assumes `docker login` has been performed, if relevant.
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerRemoteTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerRemoteTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerRemoteTagCheck {
+  docker manifest inspect $1 > /dev/null 2> /dev/null
+}
+
+# ViashDockerLocalTagCheck: check whether a Docker image is available locally
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   docker pull python:latest
+#   ViashDockerLocalTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerLocalTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerLocalTagCheck {
+  [ -n "$(docker images -q $1)" ]
+}
+
+# ViashDockerPull: pull a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPull python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPull sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPull {
+  ViashNotice "Checking if Docker image is available at '$1'"
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker pull $1 && return 0 || return 1
+  else
+    local save=$-; set +e
+    docker pull $1 2> /dev/null > /dev/null
+    local out=$?
+    [[ $save =~ e ]] && set -e
+    if [ $out -ne 0 ]; then
+      ViashWarning "Could not pull from '$1'. Docker image doesn't exist or is not accessible."
+    fi
+    return $out
+  fi
+}
+
+# ViashDockerPush: push a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPush python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPush sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPush {
+  ViashNotice "Pushing image to '$1'"
+  local save=$-; set +e
+  local out
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker push $1
+    out=$?
+  else
+    docker push $1 2> /dev/null > /dev/null
+    out=$?
+  fi
+  [[ $save =~ e ]] && set -e
+  if [ $out -eq 0 ]; then
+    ViashNotice "Container '$1' push succeeded."
+  else
+    ViashError "Container '$1' push errored. You might not be logged in or have the necessary permissions."
+  fi
+  return $out
+}
+
+# ViashDockerPullElseBuild: pull a Docker image, else build it
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# ViashDockerBuild    : a Bash function which builds a docker image, takes image identifier as argument.
+# examples:
+#   ViashDockerPullElseBuild mynewcomponent
+function ViashDockerPullElseBuild {
+  local save=$-; set +e
+  ViashDockerPull $1
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashDockerBuild $@
+  fi
+}
+
+# ViashDockerSetup: create a Docker image, according to specified docker setup strategy
+#
+# $1          : image identifier with format `[registry/]image[:tag]`
+# $2          : docker setup strategy, see DockerSetupStrategy.scala
+# examples:
+#   ViashDockerSetup mynewcomponent alwaysbuild
+function ViashDockerSetup {
+  local image_id="$1"
+  local setup_strategy="$2"
+  if [ "$setup_strategy" == "alwaysbuild" -o "$setup_strategy" == "build" -o "$setup_strategy" == "b" ]; then
+    ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspull" -o "$setup_strategy" == "pull" -o "$setup_strategy" == "p" ]; then
+    ViashDockerPull $image_id
+  elif [ "$setup_strategy" == "alwayspullelsebuild" -o "$setup_strategy" == "pullelsebuild" ]; then
+    ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspullelsecachedbuild" -o "$setup_strategy" == "pullelsecachedbuild" ]; then
+    ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayscachedbuild" -o "$setup_strategy" == "cachedbuild" -o "$setup_strategy" == "cb" ]; then
+    ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [[ "$setup_strategy" =~ ^ifneedbe ]]; then
+    local save=$-; set +e
+    ViashDockerLocalTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashInfo "Image $image_id already exists"
+    elif [ "$setup_strategy" == "ifneedbebuild" ]; then
+      ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbecachedbuild" ]; then
+      ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepull" ]; then
+      ViashDockerPull $image_id
+    elif [ "$setup_strategy" == "ifneedbepullelsebuild" ]; then
+      ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepullelsecachedbuild" ]; then
+      ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    else
+      ViashError "Unrecognised Docker strategy: $setup_strategy"
+      exit 1
+    fi
+  elif [ "$setup_strategy" == "push" -o "$setup_strategy" == "forcepush" -o "$setup_strategy" == "alwayspush" ]; then
+    ViashDockerPush "$image_id"
+  elif [ "$setup_strategy" == "pushifnotpresent" -o "$setup_strategy" == "gentlepush" -o "$setup_strategy" == "maybepush" ]; then
+    local save=$-; set +e
+    ViashDockerRemoteTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashNotice "Container '$image_id' exists, doing nothing."
+    else
+      ViashNotice "Container '$image_id' does not yet exist."
+      ViashDockerPush "$image_id"
+    fi
+  elif [ "$setup_strategy" == "donothing" -o "$setup_strategy" == "meh" ]; then
+    ViashNotice "Skipping setup."
+  else
+    ViashError "Unrecognised Docker strategy: $setup_strategy"
+    exit 1
+  fi
+}
+
+# ViashDockerCheckCommands: Check whether a docker container has the required commands
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# $@                  : commands to verify being present
+# examples:
+#   ViashDockerCheckCommands bash:4.0 bash ps foo
+function ViashDockerCheckCommands {
+  local image_id="$1"
+  shift 1
+  local commands="$@"
+  local save=$-; set +e
+  local missing # mark 'missing' as local in advance, otherwise the exit code of the command will be missing and always be '0'
+  missing=$(docker run --rm --entrypoint=sh "$image_id" -c "for command in $commands; do command -v \$command >/dev/null 2>&1; if [ \$? -ne 0 ]; then echo \$command; exit 1; fi; done")
+  local outCheck=$?
+  [[ $save =~ e ]] && set -e
+  if [ $outCheck -ne 0 ]; then
+  	ViashError "Docker container '$image_id' does not contain command '$missing'."
+  	exit 1
+  fi
+}
+
+# ViashDockerBuild: build a docker image
+# $1                               : image identifier with format `[registry/]image[:tag]`
+# $...                             : additional arguments to pass to docker build
+# $VIASH_META_TEMP_DIR             : temporary directory to store dockerfile & optional resources in
+# $VIASH_META_NAME                 : name of the component
+# $VIASH_META_RESOURCES_DIR        : directory containing the resources
+# $VIASH_VERBOSITY                 : verbosity level
+# exit code $?                     : whether or not the image was built successfully
+function ViashDockerBuild {
+  local image_id="$1"
+  shift 1
+
+  # create temporary directory to store dockerfile & optional resources in
+  local tmpdir=$(mktemp -d "$VIASH_META_TEMP_DIR/dockerbuild-$VIASH_META_NAME-XXXXXX")
+  local dockerfile="$tmpdir/Dockerfile"
+  function clean_up {
+    rm -rf "$tmpdir"
+  }
+  trap clean_up EXIT
+
+  # store dockerfile and resources
+  ViashDockerfile "$VIASH_ENGINE_ID" > "$dockerfile"
+
+  # generate the build command
+  local docker_build_cmd="docker build -t '$image_id' $@ '$VIASH_META_RESOURCES_DIR' -f '$dockerfile'"
+
+  # build the container
+  ViashNotice "Building container '$image_id' with Dockerfile"
+  ViashInfo "$docker_build_cmd"
+  local save=$-; set +e
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    eval $docker_build_cmd
+  else
+    eval $docker_build_cmd &> "$tmpdir/docker_build.log"
+  fi
+
+  # check exit code
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashError "Error occurred while building container '$image_id'"
+    if [ $VIASH_VERBOSITY -lt $VIASH_LOGCODE_INFO ]; then
+      ViashError "Transcript: --------------------------------"
+      cat "$tmpdir/docker_build.log"
+      ViashError "End of transcript --------------------------"
+    fi
+    exit 1
+  fi
+}
+
+######## End of helper functions for setting up Docker images for viash ########
+
+# ViashDockerFile: print the dockerfile to stdout
+# $1    : engine identifier
+# return : dockerfile required to run this component
+# examples:
+#   ViashDockerFile
+function ViashDockerfile {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    cat << 'VIASHDOCKER'
+FROM python:3.12-slim
+ENTRYPOINT []
+RUN apt-get update && \
+  DEBIAN_FRONTEND=noninteractive apt-get install -y procps && \
+  rm -rf /var/lib/apt/lists/*
+
+RUN pip install --upgrade pip && \
+  pip install --upgrade --no-cache-dir "anndata~=0.11.1" "mudata~=0.3.1" && \
+  python -c 'exec("try:\n  import awkward\nexcept ModuleNotFoundError:\n  exit(0)\nelse:  exit(1)")'
+
+LABEL org.opencontainers.image.authors="Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz"
+LABEL org.opencontainers.image.description="Companion container for running component ingestion_qc h5mu_to_qc_json"
+LABEL org.opencontainers.image.created="2025-08-11T08:32:06Z"
+LABEL org.opencontainers.image.source="https://github.com/openpipelines-bio/openpipeline_incubator"
+LABEL org.opencontainers.image.revision="dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+LABEL org.opencontainers.image.version="build_main"
+
+VIASHDOCKER
+  fi
+}
+
+# ViashDockerBuildArgs: return the arguments to pass to docker build
+# $1    : engine identifier
+# return : arguments to pass to docker build
+function ViashDockerBuildArgs {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    echo ""
+  fi
+}
+
+# ViashAbsolutePath: generate absolute path from relative path
+# borrowed from https://stackoverflow.com/a/21951256
+# $1     : relative filename
+# return : absolute path
+# examples:
+#   ViashAbsolutePath some_file.txt   # returns /path/to/some_file.txt
+#   ViashAbsolutePath /foo/bar/..     # returns /foo
+function ViashAbsolutePath {
+  local thePath
+  local parr
+  local outp
+  local len
+  if [[ ! "$1" =~ ^/ ]]; then
+    thePath="$PWD/$1"
+  else
+    thePath="$1"
+  fi
+  echo "$thePath" | (
+    IFS=/
+    read -a parr
+    declare -a outp
+    for i in "${parr[@]}"; do
+      case "$i" in
+      ''|.) continue ;;
+      ..)
+        len=${#outp[@]}
+        if ((len==0)); then
+          continue
+        else
+          unset outp[$((len-1))]
+        fi
+        ;;
+      *)
+        len=${#outp[@]}
+        outp[$len]="$i"
+      ;;
+      esac
+    done
+    echo /"${outp[*]}"
+  )
+}
+# ViashDockerAutodetectMount: auto configuring docker mounts from parameters
+# $1                             : The parameter value
+# returns                        : New parameter
+# $VIASH_DIRECTORY_MOUNTS        : Added another parameter to be passed to docker
+# $VIASH_DOCKER_AUTOMOUNT_PREFIX : The prefix to be used for the automounts
+# examples:
+#   ViashDockerAutodetectMount /path/to/bar      # returns '/viash_automount/path/to/bar'
+#   ViashDockerAutodetectMountArg /path/to/bar   # returns '--volume="/path/to:/viash_automount/path/to"'
+function ViashDockerAutodetectMount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  if [ -z "$base_name" ]; then
+    echo "$mount_target"
+  else
+    echo "$mount_target/$base_name"
+  fi
+}
+function ViashDockerAutodetectMountArg {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  ViashDebug "ViashDockerAutodetectMountArg $1 -> $mount_source -> $mount_target"
+  echo "--volume=\"$mount_source:$mount_target\""
+}
+function ViashDockerStripAutomount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  echo "${abs_path#$VIASH_DOCKER_AUTOMOUNT_PREFIX}"
+}
+# initialise variables
+VIASH_DIRECTORY_MOUNTS=()
+
+# configure default docker automount prefix if it is unset
+if [ -z "${VIASH_DOCKER_AUTOMOUNT_PREFIX+x}" ]; then
+  VIASH_DOCKER_AUTOMOUNT_PREFIX="/viash_automount"
+fi
+
+# initialise docker variables
+VIASH_DOCKER_RUN_ARGS=(-i --rm)
+
+
+# ViashHelp: Display helpful explanation about this executable
+function ViashHelp {
+  echo "h5mu_to_qc_json build_main"
+  echo ""
+  echo "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and"
+  echo "processed by the QC workflow, and generates:"
+  echo "- A JSON file that contains the combined data for the QC report"
+  echo "- A JSON file that defines the layout and structure of the QC report"
+  echo ""
+  echo "Inputs:"
+  echo "    --input"
+  echo "        type: file, required parameter, multiple values allowed, file must exist"
+  echo "        example: path/to/file.h5mu"
+  echo "        The input h5mu file(s)"
+  echo ""
+  echo "    --modality"
+  echo "        type: string"
+  echo "        default: rna"
+  echo "        The modality to use"
+  echo ""
+  echo "    --ingestion_method"
+  echo "        type: string, required parameter"
+  echo "        choices: [ cellranger_multi, xenium ]"
+  echo "        Method that was used to ingest the data - this will define the structure"
+  echo "        of the report that is generated."
+  echo ""
+  echo "    --obs_sample_id"
+  echo "        type: string"
+  echo "        default: sample_id"
+  echo "        The key in the h5mu file that contains the sample ID. If not provided,"
+  echo "        each H5MU file will be considered as a separate sample."
+  echo ""
+  echo "    --obs_total_counts"
+  echo "        type: string"
+  echo "        default: total_counts"
+  echo "        The key in the h5mu .obs field that contains the total counts."
+  echo ""
+  echo "    --obs_num_nonzero_vars"
+  echo "        type: string"
+  echo "        default: num_nonzero_vars"
+  echo "        The key in the h5mu .obs field that contains the number of nonzero vars."
+  echo ""
+  echo "    --obs_fraction_mitochondrial"
+  echo "        type: string"
+  echo "        default: fraction_mitochondrial"
+  echo "        The key in the h5mu .obs field that contains the fraction mitochondrial"
+  echo "        genes."
+  echo ""
+  echo "    --obs_fraction_ribosomal"
+  echo "        type: string"
+  echo "        default: fraction_ribosomal"
+  echo "        The key in the h5mu .obs field that contains the fraction ribosomal"
+  echo "        genes."
+  echo ""
+  echo "Outputs:"
+  echo "    --output"
+  echo "        type: file, required parameter, output, file must exist"
+  echo "        example: path/to/file.json"
+  echo "        The output JSON file"
+  echo ""
+  echo "    --output_reporting_json"
+  echo "        type: file, required parameter, output, file must exist"
+  echo "        example: path/to/file.json"
+  echo "        The output JSON file that defines the QC report"
+  echo ""
+  echo "Filtering & grouping options:"
+  echo "    --min_total_counts"
+  echo "        type: integer"
+  echo "        default: 10"
+  echo "        Minimum total counts for a cell to be included in the output"
+  echo ""
+  echo "    --min_num_nonzero_vars"
+  echo "        type: integer"
+  echo "        default: 10"
+  echo "        Minimum number of nonzero vars for a cell to be included in the output"
+  echo ""
+  echo "    --obs_metadata"
+  echo "        type: string, multiple values allowed"
+  echo "        example: donor_id;cell_type;batch;condition"
+  echo "        The metadata keys in the h5mu .obs to include in the output JSON."
+  echo ""
+  echo "Options for CellRanger reports:"
+  echo "    --obs_cellbender"
+  echo "        type: string, multiple values allowed"
+  echo "        default:"
+  echo "cellbender_background_fraction;cellbender_cell_probability;cellbender_cell_size;cellbender_droplet_efficiency"
+  echo "        The cellbender keys in the h5mu .obs to include in the output JSON"
+  echo ""
+  echo "    --uns_cellranger_metrics"
+  echo "        type: string"
+  echo "        default: metrics_cellranger"
+  echo "        The key in the h5mu file .uns that contains the cellranger metrics"
+  echo ""
+  echo "Options for Xenium reports:"
+  echo "    --obs_nucleus_area"
+  echo "        type: string"
+  echo "        default: nucleus_area"
+  echo "        The key in the h5mu .obs field that contains the nucleus area."
+  echo ""
+  echo "    --obs_cell_area"
+  echo "        type: string"
+  echo "        default: cell_area"
+  echo "        The key in the h5mu .obs field that contains the cell area."
+  echo ""
+  echo "    --obs_x_coord"
+  echo "        type: string"
+  echo "        default: x_coord"
+  echo "        The key in the h5mu .obs field that contains the x coordinate."
+  echo ""
+  echo "    --obs_y_coord"
+  echo "        type: string"
+  echo "        default: y_coord"
+  echo "        The key in the h5mu .obs field that contains the y coordinate."
+  echo ""
+  echo "    --obs_control_probe_counts"
+  echo "        type: string"
+  echo "        default: control_probe_counts"
+  echo "        The key in the h5mu .obs field that contains the number of control"
+  echo "        probes."
+  echo ""
+  echo "    --obs_control_codeword_counts"
+  echo "        type: string"
+  echo "        default: control_codeword_counts"
+  echo "        The key in the h5mu .obs field that contains the number of control"
+  echo "        codewords."
+  echo ""
+  echo "Viash built in Computational Requirements:"
+  echo "    ---cpus=INT"
+  echo "        Number of CPUs to use"
+  echo "    ---memory=STRING"
+  echo "        Amount of memory to use. Examples: 4GB, 3MiB."
+  echo ""
+  echo "Viash built in Docker:"
+  echo "    ---setup=STRATEGY"
+  echo "        Setup the docker container. Options are: alwaysbuild, alwayscachedbuild, ifneedbebuild, ifneedbecachedbuild, alwayspull, alwayspullelsebuild, alwayspullelsecachedbuild, ifneedbepull, ifneedbepullelsebuild, ifneedbepullelsecachedbuild, push, pushifnotpresent, donothing."
+  echo "        Default: ifneedbepullelsecachedbuild"
+  echo "    ---dockerfile"
+  echo "        Print the dockerfile to stdout."
+  echo "    ---docker_run_args=ARG"
+  echo "        Provide runtime arguments to Docker. See the documentation on \`docker run\` for more information."
+  echo "    ---docker_image_id"
+  echo "        Print the docker image id to stdout."
+  echo "    ---debug"
+  echo "        Enter the docker container for debugging purposes."
+  echo ""
+  echo "Viash built in Engines:"
+  echo "    ---engine=ENGINE_ID"
+  echo "        Specify the engine to use. Options are: docker, native."
+  echo "        Default: docker"
+}
+
+# initialise array
+VIASH_POSITIONAL_ARGS=''
+
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        -h|--help)
+            ViashHelp
+            exit
+            ;;
+        ---v|---verbose)
+            let "VIASH_VERBOSITY=VIASH_VERBOSITY+1"
+            shift 1
+            ;;
+        ---verbosity)
+            VIASH_VERBOSITY="$2"
+            shift 2
+            ;;
+        ---verbosity=*)
+            VIASH_VERBOSITY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        --version)
+            echo "h5mu_to_qc_json build_main"
+            exit
+            ;;
+        --input)
+            if [ -z "$VIASH_PAR_INPUT" ]; then
+              VIASH_PAR_INPUT="$2"
+            else
+              VIASH_PAR_INPUT="$VIASH_PAR_INPUT;""$2"
+            fi
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --input. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --input=*)
+            if [ -z "$VIASH_PAR_INPUT" ]; then
+              VIASH_PAR_INPUT=$(ViashRemoveFlags "$1")
+            else
+              VIASH_PAR_INPUT="$VIASH_PAR_INPUT;"$(ViashRemoveFlags "$1")
+            fi
+            shift 1
+            ;;
+        --modality)
+            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MODALITY="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --modality. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --modality=*)
+            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality=*\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MODALITY=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --ingestion_method)
+            [ -n "$VIASH_PAR_INGESTION_METHOD" ] && ViashError Bad arguments for option \'--ingestion_method\': \'$VIASH_PAR_INGESTION_METHOD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INGESTION_METHOD="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --ingestion_method. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --ingestion_method=*)
+            [ -n "$VIASH_PAR_INGESTION_METHOD" ] && ViashError Bad arguments for option \'--ingestion_method=*\': \'$VIASH_PAR_INGESTION_METHOD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INGESTION_METHOD=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_sample_id)
+            [ -n "$VIASH_PAR_OBS_SAMPLE_ID" ] && ViashError Bad arguments for option \'--obs_sample_id\': \'$VIASH_PAR_OBS_SAMPLE_ID\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_SAMPLE_ID="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_sample_id. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_sample_id=*)
+            [ -n "$VIASH_PAR_OBS_SAMPLE_ID" ] && ViashError Bad arguments for option \'--obs_sample_id=*\': \'$VIASH_PAR_OBS_SAMPLE_ID\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_SAMPLE_ID=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_total_counts)
+            [ -n "$VIASH_PAR_OBS_TOTAL_COUNTS" ] && ViashError Bad arguments for option \'--obs_total_counts\': \'$VIASH_PAR_OBS_TOTAL_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_TOTAL_COUNTS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_total_counts. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_total_counts=*)
+            [ -n "$VIASH_PAR_OBS_TOTAL_COUNTS" ] && ViashError Bad arguments for option \'--obs_total_counts=*\': \'$VIASH_PAR_OBS_TOTAL_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_TOTAL_COUNTS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_num_nonzero_vars)
+            [ -n "$VIASH_PAR_OBS_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--obs_num_nonzero_vars\': \'$VIASH_PAR_OBS_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_NUM_NONZERO_VARS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_num_nonzero_vars. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_num_nonzero_vars=*)
+            [ -n "$VIASH_PAR_OBS_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--obs_num_nonzero_vars=*\': \'$VIASH_PAR_OBS_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_NUM_NONZERO_VARS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_fraction_mitochondrial)
+            [ -n "$VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL" ] && ViashError Bad arguments for option \'--obs_fraction_mitochondrial\': \'$VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_fraction_mitochondrial. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_fraction_mitochondrial=*)
+            [ -n "$VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL" ] && ViashError Bad arguments for option \'--obs_fraction_mitochondrial=*\': \'$VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_fraction_ribosomal)
+            [ -n "$VIASH_PAR_OBS_FRACTION_RIBOSOMAL" ] && ViashError Bad arguments for option \'--obs_fraction_ribosomal\': \'$VIASH_PAR_OBS_FRACTION_RIBOSOMAL\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_FRACTION_RIBOSOMAL="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_fraction_ribosomal. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_fraction_ribosomal=*)
+            [ -n "$VIASH_PAR_OBS_FRACTION_RIBOSOMAL" ] && ViashError Bad arguments for option \'--obs_fraction_ribosomal=*\': \'$VIASH_PAR_OBS_FRACTION_RIBOSOMAL\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_FRACTION_RIBOSOMAL=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --output)
+            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --output=*)
+            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output=*\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --output_reporting_json)
+            [ -n "$VIASH_PAR_OUTPUT_REPORTING_JSON" ] && ViashError Bad arguments for option \'--output_reporting_json\': \'$VIASH_PAR_OUTPUT_REPORTING_JSON\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_REPORTING_JSON="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_reporting_json. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --output_reporting_json=*)
+            [ -n "$VIASH_PAR_OUTPUT_REPORTING_JSON" ] && ViashError Bad arguments for option \'--output_reporting_json=*\': \'$VIASH_PAR_OUTPUT_REPORTING_JSON\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_REPORTING_JSON=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --min_total_counts)
+            [ -n "$VIASH_PAR_MIN_TOTAL_COUNTS" ] && ViashError Bad arguments for option \'--min_total_counts\': \'$VIASH_PAR_MIN_TOTAL_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MIN_TOTAL_COUNTS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --min_total_counts. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --min_total_counts=*)
+            [ -n "$VIASH_PAR_MIN_TOTAL_COUNTS" ] && ViashError Bad arguments for option \'--min_total_counts=*\': \'$VIASH_PAR_MIN_TOTAL_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MIN_TOTAL_COUNTS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --min_num_nonzero_vars)
+            [ -n "$VIASH_PAR_MIN_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--min_num_nonzero_vars\': \'$VIASH_PAR_MIN_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MIN_NUM_NONZERO_VARS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --min_num_nonzero_vars. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --min_num_nonzero_vars=*)
+            [ -n "$VIASH_PAR_MIN_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--min_num_nonzero_vars=*\': \'$VIASH_PAR_MIN_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MIN_NUM_NONZERO_VARS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_metadata)
+            if [ -z "$VIASH_PAR_OBS_METADATA" ]; then
+              VIASH_PAR_OBS_METADATA="$2"
+            else
+              VIASH_PAR_OBS_METADATA="$VIASH_PAR_OBS_METADATA;""$2"
+            fi
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_metadata. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_metadata=*)
+            if [ -z "$VIASH_PAR_OBS_METADATA" ]; then
+              VIASH_PAR_OBS_METADATA=$(ViashRemoveFlags "$1")
+            else
+              VIASH_PAR_OBS_METADATA="$VIASH_PAR_OBS_METADATA;"$(ViashRemoveFlags "$1")
+            fi
+            shift 1
+            ;;
+        --obs_cellbender)
+            if [ -z "$VIASH_PAR_OBS_CELLBENDER" ]; then
+              VIASH_PAR_OBS_CELLBENDER="$2"
+            else
+              VIASH_PAR_OBS_CELLBENDER="$VIASH_PAR_OBS_CELLBENDER;""$2"
+            fi
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_cellbender. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_cellbender=*)
+            if [ -z "$VIASH_PAR_OBS_CELLBENDER" ]; then
+              VIASH_PAR_OBS_CELLBENDER=$(ViashRemoveFlags "$1")
+            else
+              VIASH_PAR_OBS_CELLBENDER="$VIASH_PAR_OBS_CELLBENDER;"$(ViashRemoveFlags "$1")
+            fi
+            shift 1
+            ;;
+        --uns_cellranger_metrics)
+            [ -n "$VIASH_PAR_UNS_CELLRANGER_METRICS" ] && ViashError Bad arguments for option \'--uns_cellranger_metrics\': \'$VIASH_PAR_UNS_CELLRANGER_METRICS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_UNS_CELLRANGER_METRICS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --uns_cellranger_metrics. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --uns_cellranger_metrics=*)
+            [ -n "$VIASH_PAR_UNS_CELLRANGER_METRICS" ] && ViashError Bad arguments for option \'--uns_cellranger_metrics=*\': \'$VIASH_PAR_UNS_CELLRANGER_METRICS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_UNS_CELLRANGER_METRICS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_nucleus_area)
+            [ -n "$VIASH_PAR_OBS_NUCLEUS_AREA" ] && ViashError Bad arguments for option \'--obs_nucleus_area\': \'$VIASH_PAR_OBS_NUCLEUS_AREA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_NUCLEUS_AREA="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_nucleus_area. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_nucleus_area=*)
+            [ -n "$VIASH_PAR_OBS_NUCLEUS_AREA" ] && ViashError Bad arguments for option \'--obs_nucleus_area=*\': \'$VIASH_PAR_OBS_NUCLEUS_AREA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_NUCLEUS_AREA=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_cell_area)
+            [ -n "$VIASH_PAR_OBS_CELL_AREA" ] && ViashError Bad arguments for option \'--obs_cell_area\': \'$VIASH_PAR_OBS_CELL_AREA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CELL_AREA="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_cell_area. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_cell_area=*)
+            [ -n "$VIASH_PAR_OBS_CELL_AREA" ] && ViashError Bad arguments for option \'--obs_cell_area=*\': \'$VIASH_PAR_OBS_CELL_AREA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CELL_AREA=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_x_coord)
+            [ -n "$VIASH_PAR_OBS_X_COORD" ] && ViashError Bad arguments for option \'--obs_x_coord\': \'$VIASH_PAR_OBS_X_COORD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_X_COORD="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_x_coord. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_x_coord=*)
+            [ -n "$VIASH_PAR_OBS_X_COORD" ] && ViashError Bad arguments for option \'--obs_x_coord=*\': \'$VIASH_PAR_OBS_X_COORD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_X_COORD=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_y_coord)
+            [ -n "$VIASH_PAR_OBS_Y_COORD" ] && ViashError Bad arguments for option \'--obs_y_coord\': \'$VIASH_PAR_OBS_Y_COORD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_Y_COORD="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_y_coord. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_y_coord=*)
+            [ -n "$VIASH_PAR_OBS_Y_COORD" ] && ViashError Bad arguments for option \'--obs_y_coord=*\': \'$VIASH_PAR_OBS_Y_COORD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_Y_COORD=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_control_probe_counts)
+            [ -n "$VIASH_PAR_OBS_CONTROL_PROBE_COUNTS" ] && ViashError Bad arguments for option \'--obs_control_probe_counts\': \'$VIASH_PAR_OBS_CONTROL_PROBE_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CONTROL_PROBE_COUNTS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_control_probe_counts. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_control_probe_counts=*)
+            [ -n "$VIASH_PAR_OBS_CONTROL_PROBE_COUNTS" ] && ViashError Bad arguments for option \'--obs_control_probe_counts=*\': \'$VIASH_PAR_OBS_CONTROL_PROBE_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CONTROL_PROBE_COUNTS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --obs_control_codeword_counts)
+            [ -n "$VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS" ] && ViashError Bad arguments for option \'--obs_control_codeword_counts\': \'$VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --obs_control_codeword_counts. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --obs_control_codeword_counts=*)
+            [ -n "$VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS" ] && ViashError Bad arguments for option \'--obs_control_codeword_counts=*\': \'$VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---engine)
+            VIASH_ENGINE_ID="$2"
+            shift 2
+            ;;
+        ---engine=*)
+            VIASH_ENGINE_ID="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---setup)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$2"
+            shift 2
+            ;;
+        ---setup=*)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---dockerfile)
+            VIASH_MODE='dockerfile'
+            shift 1
+            ;;
+        ---docker_run_args)
+            VIASH_DOCKER_RUN_ARGS+=("$2")
+            shift 2
+            ;;
+        ---docker_run_args=*)
+            VIASH_DOCKER_RUN_ARGS+=("$(ViashRemoveFlags "$1")")
+            shift 1
+            ;;
+        ---docker_image_id)
+            VIASH_MODE='docker_image_id'
+            shift 1
+            ;;
+        ---debug)
+            VIASH_MODE='debug'
+            shift 1
+            ;;
+        ---cpus)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---cpus. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---cpus=*)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus=*\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---memory)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---memory. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---memory=*)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory=*\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        *)  # positional arg or unknown option
+            # since the positional args will be eval'd, can we always quote, instead of using ViashQuote
+            VIASH_POSITIONAL_ARGS="$VIASH_POSITIONAL_ARGS '$1'"
+            [[ $1 == -* ]] && ViashWarning $1 looks like a parameter but is not a defined parameter and will instead be treated as a positional argument. Use "--help" to get more information on the parameters.
+            shift # past argument
+            ;;
+    esac
+done
+
+# parse positional parameters
+eval set -- $VIASH_POSITIONAL_ARGS
+
+
+if   [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  VIASH_ENGINE_TYPE='native'
+elif   [ "$VIASH_ENGINE_ID" == "docker" ]  ; then
+  VIASH_ENGINE_TYPE='docker'
+else
+  ViashError "Engine '$VIASH_ENGINE_ID' is not recognized. Options are: docker, native."
+  exit 1
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # check if docker is installed properly
+  ViashDockerInstallationCheck
+
+  # determine docker image id
+  if [[ "$VIASH_ENGINE_ID" == 'docker' ]]; then
+    VIASH_DOCKER_IMAGE_ID='images.viash-hub.com/vsh/openpipeline_incubator/ingestion_qc/h5mu_to_qc_json:build_main'
+  fi
+
+  # print dockerfile
+  if [ "$VIASH_MODE" == "dockerfile" ]; then
+    ViashDockerfile "$VIASH_ENGINE_ID"
+    exit 0
+
+  elif [ "$VIASH_MODE" == "docker_image_id" ]; then
+    echo "$VIASH_DOCKER_IMAGE_ID"
+    exit 0
+  
+  # enter docker container
+  elif [[ "$VIASH_MODE" == "debug" ]]; then
+    VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} -v '$(pwd)':/pwd --workdir /pwd -t $VIASH_DOCKER_IMAGE_ID"
+    ViashNotice "+ $VIASH_CMD"
+    eval $VIASH_CMD
+    exit 
+
+  # build docker image
+  elif [ "$VIASH_MODE" == "setup" ]; then
+    ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" "$VIASH_SETUP_STRATEGY"
+    ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+    exit 0
+  fi
+
+  # check if docker image exists
+  ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" ifneedbepullelsecachedbuild
+  ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+fi
+
+# setting computational defaults
+
+# helper function for parsing memory strings
+function ViashMemoryAsBytes {
+  local memory=`echo "$1" | tr '[:upper:]' '[:lower:]' | tr -d '[:space:]'`
+  local memory_regex='^([0-9]+)([kmgtp]i?b?|b)$'
+  if [[ $memory =~ $memory_regex ]]; then
+    local number=${memory/[^0-9]*/}
+    local symbol=${memory/*[0-9]/}
+    
+    case $symbol in
+      b)      memory_b=$number ;;
+      kb|k)   memory_b=$(( $number * 1000 )) ;;
+      mb|m)   memory_b=$(( $number * 1000 * 1000 )) ;;
+      gb|g)   memory_b=$(( $number * 1000 * 1000 * 1000 )) ;;
+      tb|t)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 )) ;;
+      pb|p)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 * 1000 )) ;;
+      kib|ki)   memory_b=$(( $number * 1024 )) ;;
+      mib|mi)   memory_b=$(( $number * 1024 * 1024 )) ;;
+      gib|gi)   memory_b=$(( $number * 1024 * 1024 * 1024 )) ;;
+      tib|ti)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 )) ;;
+      pib|pi)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 * 1024 )) ;;
+    esac
+    echo "$memory_b"
+  fi
+}
+# compute memory in different units
+if [ ! -z ${VIASH_META_MEMORY+x} ]; then
+  VIASH_META_MEMORY_B=`ViashMemoryAsBytes $VIASH_META_MEMORY`
+  # do not define other variables if memory_b is an empty string
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_META_MEMORY_KB=$(( ($VIASH_META_MEMORY_B+999) / 1000 ))
+    VIASH_META_MEMORY_MB=$(( ($VIASH_META_MEMORY_KB+999) / 1000 ))
+    VIASH_META_MEMORY_GB=$(( ($VIASH_META_MEMORY_MB+999) / 1000 ))
+    VIASH_META_MEMORY_TB=$(( ($VIASH_META_MEMORY_GB+999) / 1000 ))
+    VIASH_META_MEMORY_PB=$(( ($VIASH_META_MEMORY_TB+999) / 1000 ))
+    VIASH_META_MEMORY_KIB=$(( ($VIASH_META_MEMORY_B+1023) / 1024 ))
+    VIASH_META_MEMORY_MIB=$(( ($VIASH_META_MEMORY_KIB+1023) / 1024 ))
+    VIASH_META_MEMORY_GIB=$(( ($VIASH_META_MEMORY_MIB+1023) / 1024 ))
+    VIASH_META_MEMORY_TIB=$(( ($VIASH_META_MEMORY_GIB+1023) / 1024 ))
+    VIASH_META_MEMORY_PIB=$(( ($VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  else
+    # unset memory if string is empty
+    unset $VIASH_META_MEMORY_B
+  fi
+fi
+# unset nproc if string is empty
+if [ -z "$VIASH_META_CPUS" ]; then
+  unset $VIASH_META_CPUS
+fi
+
+
+# check whether required parameters exist
+if [ -z ${VIASH_PAR_INPUT+x} ]; then
+  ViashError '--input' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_INGESTION_METHOD+x} ]; then
+  ViashError '--ingestion_method' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_OUTPUT+x} ]; then
+  ViashError '--output' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_OUTPUT_REPORTING_JSON+x} ]; then
+  ViashError '--output_reporting_json' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_NAME+x} ]; then
+  ViashError 'name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then
+  ViashError 'functionality_name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_RESOURCES_DIR+x} ]; then
+  ViashError 'resources_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_EXECUTABLE+x} ]; then
+  ViashError 'executable' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_CONFIG+x} ]; then
+  ViashError 'config' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_TEMP_DIR+x} ]; then
+  ViashError 'temp_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+
+# filling in defaults
+if [ -z ${VIASH_PAR_MODALITY+x} ]; then
+  VIASH_PAR_MODALITY="rna"
+fi
+if [ -z ${VIASH_PAR_OBS_SAMPLE_ID+x} ]; then
+  VIASH_PAR_OBS_SAMPLE_ID="sample_id"
+fi
+if [ -z ${VIASH_PAR_OBS_TOTAL_COUNTS+x} ]; then
+  VIASH_PAR_OBS_TOTAL_COUNTS="total_counts"
+fi
+if [ -z ${VIASH_PAR_OBS_NUM_NONZERO_VARS+x} ]; then
+  VIASH_PAR_OBS_NUM_NONZERO_VARS="num_nonzero_vars"
+fi
+if [ -z ${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL+x} ]; then
+  VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL="fraction_mitochondrial"
+fi
+if [ -z ${VIASH_PAR_OBS_FRACTION_RIBOSOMAL+x} ]; then
+  VIASH_PAR_OBS_FRACTION_RIBOSOMAL="fraction_ribosomal"
+fi
+if [ -z ${VIASH_PAR_MIN_TOTAL_COUNTS+x} ]; then
+  VIASH_PAR_MIN_TOTAL_COUNTS="10"
+fi
+if [ -z ${VIASH_PAR_MIN_NUM_NONZERO_VARS+x} ]; then
+  VIASH_PAR_MIN_NUM_NONZERO_VARS="10"
+fi
+if [ -z ${VIASH_PAR_OBS_CELLBENDER+x} ]; then
+  VIASH_PAR_OBS_CELLBENDER="cellbender_background_fraction;cellbender_cell_probability;cellbender_cell_size;cellbender_droplet_efficiency"
+fi
+if [ -z ${VIASH_PAR_UNS_CELLRANGER_METRICS+x} ]; then
+  VIASH_PAR_UNS_CELLRANGER_METRICS="metrics_cellranger"
+fi
+if [ -z ${VIASH_PAR_OBS_NUCLEUS_AREA+x} ]; then
+  VIASH_PAR_OBS_NUCLEUS_AREA="nucleus_area"
+fi
+if [ -z ${VIASH_PAR_OBS_CELL_AREA+x} ]; then
+  VIASH_PAR_OBS_CELL_AREA="cell_area"
+fi
+if [ -z ${VIASH_PAR_OBS_X_COORD+x} ]; then
+  VIASH_PAR_OBS_X_COORD="x_coord"
+fi
+if [ -z ${VIASH_PAR_OBS_Y_COORD+x} ]; then
+  VIASH_PAR_OBS_Y_COORD="y_coord"
+fi
+if [ -z ${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS+x} ]; then
+  VIASH_PAR_OBS_CONTROL_PROBE_COUNTS="control_probe_counts"
+fi
+if [ -z ${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS+x} ]; then
+  VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS="control_codeword_counts"
+fi
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_INPUT" ]; then
+  IFS=';'
+  set -f
+  for file in $VIASH_PAR_INPUT; do
+    unset IFS
+    if [ ! -e "$file" ]; then
+      ViashError "Input file '$file' does not exist."
+      exit 1
+    fi
+  done
+  set +f
+fi
+
+# check whether parameters values are of the right type
+if [[ -n "$VIASH_PAR_MIN_TOTAL_COUNTS" ]]; then
+  if ! [[ "$VIASH_PAR_MIN_TOTAL_COUNTS" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError '--min_total_counts' has to be an integer. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_PAR_MIN_NUM_NONZERO_VARS" ]]; then
+  if ! [[ "$VIASH_PAR_MIN_NUM_NONZERO_VARS" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError '--min_num_nonzero_vars' has to be an integer. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_CPUS" ]]; then
+  if ! [[ "$VIASH_META_CPUS" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'cpus' has to be an integer. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_B" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_B" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_b' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+
+# check whether value is belongs to a set of choices
+if [ ! -z "$VIASH_PAR_INGESTION_METHOD" ]; then
+  VIASH_PAR_INGESTION_METHOD_CHOICES=("cellranger_multi;xenium")
+  IFS=';'
+  set -f
+  if ! [[ ";${VIASH_PAR_INGESTION_METHOD_CHOICES[*]};" =~ ";$VIASH_PAR_INGESTION_METHOD;" ]]; then
+    ViashError '--ingestion_method' specified value of \'$VIASH_PAR_INGESTION_METHOD\' is not in the list of allowed values. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+  set +f
+  unset IFS
+fi
+
+# create parent directories of output files, if so desired
+if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -d "$(dirname "$VIASH_PAR_OUTPUT")" ]; then
+  mkdir -p "$(dirname "$VIASH_PAR_OUTPUT")"
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT_REPORTING_JSON" ] && [ ! -d "$(dirname "$VIASH_PAR_OUTPUT_REPORTING_JSON")" ]; then
+  mkdir -p "$(dirname "$VIASH_PAR_OUTPUT_REPORTING_JSON")"
+fi
+
+if  [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  if [ "$VIASH_MODE" == "run" ]; then
+    VIASH_CMD="bash"
+  else
+    ViashError "Engine '$VIASH_ENGINE_ID' does not support mode '$VIASH_MODE'."
+    exit 1
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # detect volumes from file arguments
+  VIASH_CHOWN_VARS=()
+if [ ! -z "$VIASH_PAR_INPUT" ]; then
+  VIASH_TEST_INPUT=()
+  IFS=';'
+  for var in $VIASH_PAR_INPUT; do
+    unset IFS
+    VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$var")" )
+    var=$(ViashDockerAutodetectMount "$var")
+    VIASH_TEST_INPUT+=( "$var" )
+  done
+  VIASH_PAR_INPUT=$(IFS=';' ; echo "${VIASH_TEST_INPUT[*]}")
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_OUTPUT")" )
+  VIASH_PAR_OUTPUT=$(ViashDockerAutodetectMount "$VIASH_PAR_OUTPUT")
+  VIASH_CHOWN_VARS+=( "$VIASH_PAR_OUTPUT" )
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT_REPORTING_JSON" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_OUTPUT_REPORTING_JSON")" )
+  VIASH_PAR_OUTPUT_REPORTING_JSON=$(ViashDockerAutodetectMount "$VIASH_PAR_OUTPUT_REPORTING_JSON")
+  VIASH_CHOWN_VARS+=( "$VIASH_PAR_OUTPUT_REPORTING_JSON" )
+fi
+if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_RESOURCES_DIR")" )
+  VIASH_META_RESOURCES_DIR=$(ViashDockerAutodetectMount "$VIASH_META_RESOURCES_DIR")
+fi
+if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_EXECUTABLE")" )
+  VIASH_META_EXECUTABLE=$(ViashDockerAutodetectMount "$VIASH_META_EXECUTABLE")
+fi
+if [ ! -z "$VIASH_META_CONFIG" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_CONFIG")" )
+  VIASH_META_CONFIG=$(ViashDockerAutodetectMount "$VIASH_META_CONFIG")
+fi
+if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_TEMP_DIR")" )
+  VIASH_META_TEMP_DIR=$(ViashDockerAutodetectMount "$VIASH_META_TEMP_DIR")
+fi
+  
+  # get unique mounts
+  VIASH_UNIQUE_MOUNTS=($(for val in "${VIASH_DIRECTORY_MOUNTS[@]}"; do echo "$val"; done | sort -u))
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # change file ownership
+  function ViashPerformChown {
+    if (( ${#VIASH_CHOWN_VARS[@]} )); then
+      set +e
+      VIASH_CMD="docker run --entrypoint=bash --rm ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID -c 'chown $(id -u):$(id -g) --silent --recursive ${VIASH_CHOWN_VARS[@]}'"
+      ViashDebug "+ $VIASH_CMD"
+      eval $VIASH_CMD
+      set -e
+    fi
+  }
+  trap ViashPerformChown EXIT
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # helper function for filling in extra docker args
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--memory=${VIASH_META_MEMORY_B}")
+  fi
+  if [ ! -z "$VIASH_META_CPUS" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--cpus=${VIASH_META_CPUS}")
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID"
+fi
+
+
+# set dependency paths
+
+
+ViashDebug "Running command: $(echo $VIASH_CMD)"
+cat << VIASHEOF | eval $VIASH_CMD
+set -e
+tempscript=\$(mktemp "$VIASH_META_TEMP_DIR/viash-run-h5mu_to_qc_json-XXXXXX").py
+function clean_up {
+  rm "\$tempscript"
+}
+function interrupt {
+  echo -e "\nCTRL-C Pressed..."
+  exit 1
+}
+trap clean_up EXIT
+trap interrupt INT SIGINT
+cat > "\$tempscript" << 'VIASHMAIN'
+import json
+import pandas as pd
+from pathlib import Path
+import anndata as ad
+import h5py
+import sys
+import os
+import shutil
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\'/\'\"\'\"r\'}'.split(';')"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'ingestion_method': $( if [ ! -z ${VIASH_PAR_INGESTION_METHOD+x} ]; then echo "r'${VIASH_PAR_INGESTION_METHOD//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_sample_id': $( if [ ! -z ${VIASH_PAR_OBS_SAMPLE_ID+x} ]; then echo "r'${VIASH_PAR_OBS_SAMPLE_ID//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_total_counts': $( if [ ! -z ${VIASH_PAR_OBS_TOTAL_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_TOTAL_COUNTS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OBS_NUM_NONZERO_VARS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_fraction_mitochondrial': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_fraction_ribosomal': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_RIBOSOMAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_RIBOSOMAL//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'output_reporting_json': $( if [ ! -z ${VIASH_PAR_OUTPUT_REPORTING_JSON+x} ]; then echo "r'${VIASH_PAR_OUTPUT_REPORTING_JSON//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'min_total_counts': $( if [ ! -z ${VIASH_PAR_MIN_TOTAL_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MIN_TOTAL_COUNTS//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'min_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_MIN_NUM_NONZERO_VARS+x} ]; then echo "int(r'${VIASH_PAR_MIN_NUM_NONZERO_VARS//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'obs_metadata': $( if [ ! -z ${VIASH_PAR_OBS_METADATA+x} ]; then echo "r'${VIASH_PAR_OBS_METADATA//\'/\'\"\'\"r\'}'.split(';')"; else echo None; fi ),
+  'obs_cellbender': $( if [ ! -z ${VIASH_PAR_OBS_CELLBENDER+x} ]; then echo "r'${VIASH_PAR_OBS_CELLBENDER//\'/\'\"\'\"r\'}'.split(';')"; else echo None; fi ),
+  'uns_cellranger_metrics': $( if [ ! -z ${VIASH_PAR_UNS_CELLRANGER_METRICS+x} ]; then echo "r'${VIASH_PAR_UNS_CELLRANGER_METRICS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_nucleus_area': $( if [ ! -z ${VIASH_PAR_OBS_NUCLEUS_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_NUCLEUS_AREA//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_cell_area': $( if [ ! -z ${VIASH_PAR_OBS_CELL_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_AREA//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_x_coord': $( if [ ! -z ${VIASH_PAR_OBS_X_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_X_COORD//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_y_coord': $( if [ ! -z ${VIASH_PAR_OBS_Y_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_Y_COORD//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_control_probe_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'obs_control_codeword_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS//\'/\'\"\'\"r\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\'/\'\"\'\"r\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+par["obs_cellbender"] = {} if not par["obs_cellbender"] else par["obs_cellbender"]
+
+
+def transform_df(df):
+    """Transform a DataFrame into the annotation object format."""
+    columns = []
+    for name in df.columns:
+        data = df[name]
+
+        # Determine dtype
+        if pd.api.types.is_integer_dtype(data):
+            dtype = "integer"
+        elif pd.api.types.is_float_dtype(data):
+            dtype = "numeric"
+        elif pd.api.types.is_categorical_dtype(data):
+            dtype = "categorical"
+        else:
+            raise ValueError(f"Unknown/unsupported data type for column {name}")
+
+        column_info = {"name": name, "dtype": dtype}
+
+        if dtype == "categorical":
+            column_info["data"] = data.cat.codes.tolist()
+            column_info["categories"] = data.cat.categories.tolist()
+        else:
+            column_info["data"] = [None if pd.isna(x) else x for x in data]
+
+        columns.append(column_info)
+
+    return {"num_rows": len(df), "num_cols": len(df.columns), "columns": columns}
+
+
+def check_optional_obs_keys(obs, keys, message):
+    missing_keys = [key for key in keys if key not in obs.columns]
+    if missing_keys:
+        logger.info(f"Missing keys in obs: {', '.join(missing_keys)}. {message}")
+
+
+def transform_cellranger_metrics(uns, sample_id):
+    if not par["uns_cellranger_metrics"] in uns:
+        raise ValueError(f"Could not find cellranger metrics in uns: {par['uns_cellranger_metrics']}. Provide correct value for --uns_cellranger_metrics or make sure data was ingested using CellRanger multi.")
+
+    cellranger_metrics = (
+        uns[par["uns_cellranger_metrics"]]
+        .pivot_table(
+            index=[],
+            columns="Metric Name",
+            values="Metric Value",
+            aggfunc="first",
+        )
+        .reset_index(drop=True)
+    )
+
+    cellranger_metrics.columns.name = None
+    # Remove thousands separator and convert to numeric
+    cellranger_metrics = cellranger_metrics.map(
+        lambda x: (
+            pd.to_numeric(x.replace(",", ""), errors="coerce")
+            if isinstance(x, str)
+            else x
+        )
+    )
+    # Replace spaces with underscores in column names
+    cellranger_metrics.columns = cellranger_metrics.columns.str.replace(" ", "_")
+    for col in cellranger_metrics.columns:
+        cellranger_metrics[col] = pd.to_numeric(cellranger_metrics[col], errors="coerce")
+    cellranger_metrics["sample_id"] = [sample_id[0]]
+
+    return cellranger_metrics
+
+
+def format_cellbender_columns(mod_obs):
+    # Check if celbender was run on the dataset
+    if par["obs_cellbender"]:
+        check_optional_obs_keys(mod_obs, par["obs_cellbender"], "Run cellbender first to include these metrics.")
+
+    cellbender_obs_keys = [column for column in par["obs_cellbender"] if column in mod_obs]
+
+    for key in cellbender_obs_keys:
+        if not pd.api.types.is_float_dtype(mod_obs[key]):
+            try:
+                mod_obs[key] = mod_obs[key].astype("float16")
+            except ValueError:
+                raise ValueError(f"Could not convert column {key} to a float dtype. Please make sure all cellbender metrics are numeric.")
+
+    return cellbender_obs_keys, mod_obs
+
+
+def format_required_columns(required_keys, mod_obs):
+
+    for key in required_keys:
+        if not pd.api.types.is_numeric_dtype(mod_obs[key]):
+            raise ValueError(f"Column {key} must be a numeric dtype.")
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_total_counts"]]):
+        logger.info(f"Converting {par['obs_total_counts']} from {mod_obs[par['obs_total_counts']].dtype} to integer dtype...")
+        mod_obs[par["obs_total_counts"]] = mod_obs[par["obs_total_counts"]].astype(int)
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_num_nonzero_vars"]]):
+        logger.info(f"Converting {par['obs_num_nonzero_vars']} from {mod_obs[par['obs_num_nonzero_vars']].dtype} to integer dtype...")
+        mod_obs[par["obs_num_nonzero_vars"]] = mod_obs[par["obs_num_nonzero_vars"]].astype(int)
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_mitochondrial"]]):
+        logger.info(f"Converting {par['obs_fraction_mitochondrial']} from {mod_obs[par['obs_fraction_mitochondrial']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_mitochondrial"]] = mod_obs[par["obs_fraction_mitochondrial"]].astype("float16")
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_ribosomal"]]):
+        logger.info(f"Converting {par['obs_fraction_ribosomal']} from {mod_obs[par['obs_fraction_ribosomal']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_ribosomal"]] = mod_obs[par["obs_fraction_ribosomal"]].astype("float16")
+
+    return mod_obs
+
+
+def format_categorical_columns(mod_obs):
+    # Fetch all categorical columns for grouping if no columns are provided
+    if not par["obs_metadata"]:
+        metadata_obs_keys = mod_obs.select_dtypes(include=["object", "category"]).columns.tolist()
+        if par["obs_sample_id"] in metadata_obs_keys:
+            metadata_obs_keys.remove(par["obs_sample_id"])
+    else:
+        check_optional_obs_keys(mod_obs, par["obs_metadata"], "Make sure requested metadata colmuns are present in obs.")
+        metadata_obs_keys = [key for key in par["obs_metadata"] if key in mod_obs]
+
+    for key in metadata_obs_keys:
+        if not isinstance(key, pd.CategoricalDtype):
+            logger.info(f"{key} is not a categorical dtype. Converting {key} from {mod_obs[key].dtype} to categorical dtype...")
+            mod_obs[key] = mod_obs[key].astype(str).astype("category")
+
+    return metadata_obs_keys, mod_obs
+
+
+def generate_cellranger_stats(mod_obs, uns, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Fetch and format cellbender columns
+    cellbender_obs_keys, mod_obs = format_cellbender_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in cellbender_obs_keys},
+            **{key: mod_obs[key] for key in metadata_obs_keys},
+        }
+    )
+
+    cellranger_stats = transform_cellranger_metrics(uns, sample_id)
+
+    return cell_rna_stats, cellranger_stats
+
+
+def format_xenium_columns(mod_obs):
+
+    mod_obs["nucleus_ratio"] = mod_obs[par["obs_nucleus_area"]] / mod_obs[par["obs_cell_area"]]
+
+    xenium_formatted_columns = [par["obs_cell_area"], "nucleus_ratio", "x_coord", "y_coord"]
+    for key in xenium_formatted_columns:
+        mod_obs[key] = mod_obs[key].astype("float16")
+
+    return mod_obs, xenium_formatted_columns
+
+
+def generate_xenium_stats(mod_obs, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format xenium-specific columns
+    mod_obs, xenium_formatted_columns = format_xenium_columns(mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in xenium_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+
+    return cell_rna_stats
+
+
+def concatenate_dataframes(dfs):
+    '''Concatenates a list of dataframes into a single dataframe, preserving categorical columns.'''
+    df = pd.concat(dfs, ignore_index=True)
+
+    # Find categorical columns that became object columms
+    for col in df.columns:
+        if any(df[col].dtype.name == 'category' for df in dfs if col in df.columns):
+            # Get all categorical series for this column
+            cat_series = [df[col] for df in dfs if col in df.columns and df[col].dtype.name == 'category']
+            if cat_series:
+                # Union the categories and apply to result
+                unioned = pd.api.types.union_categoricals(cat_series)
+                df[col] = pd.Categorical(df[col], categories=unioned.categories)
+    return df
+
+
+def main(par):
+    cell_stats_dfs = []
+    sample_stats_dfs = []
+    metrics_cellranger_dfs = []
+
+    for i, mudata_file in enumerate(par["input"]):
+        logger.info(f"Processing {mudata_file}")
+
+        # read h5mu file
+        file = h5py.File(mudata_file, "r")
+
+        # read the necessary info
+        grp_mod = file["mod"][par["modality"]]
+        mod_obs = ad.experimental.read_elem(grp_mod["obs"])
+        mod_obsm = ad.experimental.read_elem(grp_mod["obsm"])
+        uns = ad.experimental.read_elem(file["uns"])
+
+        # close the h5mu file
+        file.close()
+
+        barcodes_original_count = mod_obs.shape[0]
+
+        # Add coordinates to obs before filtering
+        if par["ingestion_method"] == "xenium":
+            mod_obs["x_coord"] = mod_obsm["spatial"][:, 0]
+            mod_obs["y_coord"] = mod_obsm["spatial"][:, 1]
+
+        # Pre-filter cells
+        logger.info("Pre-filtering cells based on counts...")
+        if "min_total_counts" in par:
+            mod_obs = mod_obs[mod_obs["total_counts"] >= par["min_total_counts"]]
+        if "min_num_nonzero_vars" in par:
+            mod_obs = mod_obs[mod_obs["num_nonzero_vars"] >= par["min_num_nonzero_vars"]]
+        barcodes_filtered_count = mod_obs.shape[0]
+
+        # Detect sample id's
+        logger.info("Detecting sample id's...")
+        sample_id = (
+            mod_obs[par["obs_sample_id"]].tolist()
+            if par["obs_sample_id"] in mod_obs.columns
+            else [f"sample_{i}"] * mod_obs.shape[0]
+        )
+
+        # Generating sample summary statistics
+        logger.info("Generating sample summary statistics...")
+        required_keys = [
+            par["obs_total_counts"],
+            par["obs_num_nonzero_vars"],
+            par["obs_fraction_mitochondrial"],
+            par["obs_fraction_ribosomal"]
+            ]
+        missing_keys = [key for key in required_keys if key not in mod_obs.columns]
+        if missing_keys:
+            raise ValueError(f"Missing keys in obs: {', '.join(missing_keys)}")
+
+        sample_summary = {
+            "sample_id": pd.Categorical([sample_id[0]]),
+            "rna_num_barcodes": [barcodes_original_count],
+            "rna_num_barcodes_filtered": [barcodes_filtered_count],
+            "rna_sum_total_counts": [mod_obs[par["obs_total_counts"]].sum()],
+            "rna_median_total_counts": [mod_obs[par["obs_total_counts"]].median()],
+            "rna_overall_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].sum()],
+            "rna_median_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].median()],
+        }
+
+        if par["ingestion_method"] == "xenium":
+            sample_summary["control_probe_percentage"] = mod_obs[par["obs_control_probe_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+            sample_summary["negative_decoding_percentage"] = mod_obs[par["obs_control_codeword_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+
+        sample_summary_stats = pd.DataFrame(sample_summary)
+
+        if par["ingestion_method"] == "cellranger_multi":
+            cell_rna_stats, cellranger_stats = generate_cellranger_stats(mod_obs, uns, sample_id, required_keys)
+            metrics_cellranger_dfs.append(cellranger_stats)
+
+        if par["ingestion_method"] == "xenium":
+            cell_rna_stats = generate_xenium_stats(mod_obs, sample_id, required_keys)
+
+        cell_stats_dfs.append(cell_rna_stats)
+        sample_stats_dfs.append(sample_summary_stats)
+
+    # Combine dataframes of all samples
+    logger.info("Combining data of all samples into single object...")
+    combined_cell_stats = concatenate_dataframes(cell_stats_dfs)
+    combined_sample_stats = concatenate_dataframes(sample_stats_dfs)
+    if par["ingestion_method"] == "cellranger_multi":
+        combined_metrics_cellranger = concatenate_dataframes(metrics_cellranger_dfs)
+
+    report_categories = [combined_cell_stats, combined_sample_stats]
+
+    if par["ingestion_method"] == "cellranger_multi":
+        report_categories.append(combined_metrics_cellranger)
+
+    for df in report_categories:
+        df["sample_id"] = pd.Categorical(df["sample_id"])
+
+    output = {
+        "cell_rna_stats": transform_df(combined_cell_stats),
+        "sample_summary_stats": transform_df(combined_sample_stats)
+    }
+
+    if par["ingestion_method"] == "cellranger_multi":
+        output["metrics_cellranger_stats"] = transform_df(combined_metrics_cellranger)
+
+    logger.info(f"Writing output data json to {par['output']}")
+    output_path = Path(par["output"])
+    with open(output_path, "w") as f:
+        json.dump(output, f, indent=2)
+
+    report_structures = {
+        "cellranger_multi": os.path.join(meta["resources_dir"], "report_structure/cellranger.json"),
+        "xenium": os.path.join(meta["resources_dir"], "report_structure/xenium.json")
+    }
+
+    logger.info(f"Writing output report structure json to {par['output_reporting_json']}")
+    shutil.copy(report_structures[par["ingestion_method"]], par["output_reporting_json"])
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "\$tempscript" &
+wait "\$!"
+
+VIASHEOF
+
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # strip viash automount from file paths
+  
+  if [ ! -z "$VIASH_PAR_INPUT" ]; then
+    unset VIASH_TEST_INPUT
+    IFS=';'
+    for var in $VIASH_PAR_INPUT; do
+      unset IFS
+      if [ -z "$VIASH_TEST_INPUT" ]; then
+      VIASH_TEST_INPUT="$(ViashDockerStripAutomount "$var")"
+    else
+      VIASH_TEST_INPUT="$VIASH_TEST_INPUT;""$(ViashDockerStripAutomount "$var")"
+    fi
+    done
+    VIASH_PAR_INPUT="$VIASH_TEST_INPUT"
+  fi
+  if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
+    VIASH_PAR_OUTPUT=$(ViashDockerStripAutomount "$VIASH_PAR_OUTPUT")
+  fi
+  if [ ! -z "$VIASH_PAR_OUTPUT_REPORTING_JSON" ]; then
+    VIASH_PAR_OUTPUT_REPORTING_JSON=$(ViashDockerStripAutomount "$VIASH_PAR_OUTPUT_REPORTING_JSON")
+  fi
+  if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+    VIASH_META_RESOURCES_DIR=$(ViashDockerStripAutomount "$VIASH_META_RESOURCES_DIR")
+  fi
+  if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+    VIASH_META_EXECUTABLE=$(ViashDockerStripAutomount "$VIASH_META_EXECUTABLE")
+  fi
+  if [ ! -z "$VIASH_META_CONFIG" ]; then
+    VIASH_META_CONFIG=$(ViashDockerStripAutomount "$VIASH_META_CONFIG")
+  fi
+  if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+    VIASH_META_TEMP_DIR=$(ViashDockerStripAutomount "$VIASH_META_TEMP_DIR")
+  fi
+fi
+
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -e "$VIASH_PAR_OUTPUT" ]; then
+  ViashError "Output file '$VIASH_PAR_OUTPUT' does not exist."
+  exit 1
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT_REPORTING_JSON" ] && [ ! -e "$VIASH_PAR_OUTPUT_REPORTING_JSON" ]; then
+  ViashError "Output file '$VIASH_PAR_OUTPUT_REPORTING_JSON' does not exist."
+  exit 1
+fi
+
+
+exit 0
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
new file mode 100644
index 0000000..18769ba
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
@@ -0,0 +1,162 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "SampleQC",
+      "key": "metrics_cellranger_stats",
+      "additionalAxes": false,
+      "defaultFilters": [
+        {
+          "type": "bar",
+          "field": "Number_of_reads_in_the_library",
+          "label": "Number of reads per library",
+          "description": "Sequencing depth per sample. Higher values generally indicate more comprehensive cell profiling.",
+          "nBins": 10,
+          "groupBy": "sample_id",
+          "xAxisType": "linear",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Confidently_mapped_reads_in_cells",
+          "label": "Confidently mapped reads in cells",
+          "description": "Number of reads that were mapped unambiguously to the reference genome within cell-containing droplets.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Estimated_number_of_cells",
+          "label": "Estimated number of cells",
+          "description": "CellRanger's estimate of the number of cells per sample based on the UMI count distribution.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Sequencing_saturation",
+          "label": "Sequencing saturation",
+          "description": "Fraction of reads that are duplicates of existing UMIs. Higher values suggest deeper sequencing coverage.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        }
+      ]
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "pct_of_counts_in_top_50_vars",
+          "label": "Fraction UMI in top 50 genes per cell",
+          "description": "Proportion of RNA molecules from the 50 most-expressed genes in each cell.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_probability",
+          "label": "CellBender cell probability",
+          "description": "CellBender's statistical confidence (0-1) that a barcode represents a real cell, with higher values indicating stronger confidence.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_background_fraction",
+          "label": "CellBender background fraction",
+          "description": "Estimated percentage of each cell's RNA that comes from the ambient solution rather than the cell itself.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_size",
+          "label": "CellBender cell size",
+          "description": "CellBender's estimate of the true number of RNA molecules in each cell after removing ambient contamination. Reflects actual cell RNA content rather than raw UMI counts.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_droplet_efficiency",
+          "label": "CellBender droplet efficiency",
+          "description": "CellBender's estimate of how efficiently each droplet captured RNA molecules. Higher values indicate more reliable RNA sampling within individual droplets.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
new file mode 100644
index 0000000..9dc7a8b
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
@@ -0,0 +1,91 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "cell_area",
+          "label": "Segmented cell area",
+          "description": "Area of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "nucleus_ratio",
+          "label": "Nucleus Ratio",
+          "description": "Ratio of the nucleus area to the segmented cell area.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
\ No newline at end of file
diff --git a/target/_private/executable/ingestion_qc/h5mu_to_qc_json/setup_logger.py b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/_private/executable/ingestion_qc/h5mu_to_qc_json/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
new file mode 100644
index 0000000..6b67894
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
@@ -0,0 +1,483 @@
+name: "h5mu_to_qc_json"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu file(s)"
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "The modality to use"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ingestion_method"
+    description: "Method that was used to ingest the data - this will define the structure\
+      \ of the report that is generated."
+    info: null
+    required: true
+    choices:
+    - "cellranger_multi"
+    - "xenium"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_sample_id"
+    description: "The key in the h5mu file that contains the sample ID. If not provided,\
+      \ each H5MU file will be considered as a separate sample."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_total_counts"
+    description: "The key in the h5mu .obs field that contains the total counts."
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_num_nonzero_vars"
+    description: "The key in the h5mu .obs field that contains the number of nonzero\
+      \ vars."
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_mitochondrial"
+    description: "The key in the h5mu .obs field that contains the fraction mitochondrial\
+      \ genes."
+    info: null
+    default:
+    - "fraction_mitochondrial"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_ribosomal"
+    description: "The key in the h5mu .obs field that contains the fraction ribosomal\
+      \ genes."
+    info: null
+    default:
+    - "fraction_ribosomal"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The output JSON file"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_reporting_json"
+    description: "The output JSON file that defines the QC report"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering & grouping options"
+  arguments:
+  - type: "integer"
+    name: "--min_total_counts"
+    description: "Minimum total counts for a cell to be included in the output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_num_nonzero_vars"
+    description: "Minimum number of nonzero vars for a cell to be included in the\
+      \ output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_metadata"
+    description: "The metadata keys in the h5mu .obs to include in the output JSON."
+    info: null
+    example:
+    - "donor_id;cell_type;batch;condition"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Options for CellRanger reports"
+  arguments:
+  - type: "string"
+    name: "--obs_cellbender"
+    description: "The cellbender keys in the h5mu .obs to include in the output JSON"
+    info: null
+    default:
+    - "cellbender_background_fraction"
+    - "cellbender_cell_probability"
+    - "cellbender_cell_size"
+    - "cellbender_droplet_efficiency"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_cellranger_metrics"
+    description: "The key in the h5mu file .uns that contains the cellranger metrics"
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options for Xenium reports"
+  arguments:
+  - type: "string"
+    name: "--obs_nucleus_area"
+    description: "The key in the h5mu .obs field that contains the nucleus area."
+    info: null
+    default:
+    - "nucleus_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_area"
+    description: "The key in the h5mu .obs field that contains the cell area."
+    info: null
+    default:
+    - "cell_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_x_coord"
+    description: "The key in the h5mu .obs field that contains the x coordinate."
+    info: null
+    default:
+    - "x_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_y_coord"
+    description: "The key in the h5mu .obs field that contains the y coordinate."
+    info: null
+    default:
+    - "y_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_probe_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ probes."
+    info: null
+    default:
+    - "control_probe_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_codeword_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ codewords."
+    info: null
+    default:
+    - "control_codeword_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "report_structure"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx\
+  \ and processed by the QC workflow, and generates:\n- A JSON file that contains\
+  \ the combined data for the QC report\n- A JSON file that defines the layout and\
+  \ structure of the QC report\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "resources_test"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json"
+  executable: "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf
new file mode 100644
index 0000000..bc43249
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf
@@ -0,0 +1,4573 @@
+// h5mu_to_qc_json build_main
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "h5mu_to_qc_json",
+  "namespace" : "ingestion_qc",
+  "version" : "build_main",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The input h5mu file(s)",
+          "example" : [
+            "path/to/file.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "The modality to use",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ingestion_method",
+          "description" : "Method that was used to ingest the data - this will define the structure of the report that is generated.",
+          "required" : true,
+          "choices" : [
+            "cellranger_multi",
+            "xenium"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_sample_id",
+          "description" : "The key in the h5mu file that contains the sample ID. If not provided, each H5MU file will be considered as a separate sample.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_total_counts",
+          "description" : "The key in the h5mu .obs field that contains the total counts.",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_num_nonzero_vars",
+          "description" : "The key in the h5mu .obs field that contains the number of nonzero vars.",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_fraction_mitochondrial",
+          "description" : "The key in the h5mu .obs field that contains the fraction mitochondrial genes.",
+          "default" : [
+            "fraction_mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_fraction_ribosomal",
+          "description" : "The key in the h5mu .obs field that contains the fraction ribosomal genes.",
+          "default" : [
+            "fraction_ribosomal"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The output JSON file",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_reporting_json",
+          "description" : "The output JSON file that defines the QC report",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering & grouping options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_total_counts",
+          "description" : "Minimum total counts for a cell to be included in the output",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_num_nonzero_vars",
+          "description" : "Minimum number of nonzero vars for a cell to be included in the output",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_metadata",
+          "description" : "The metadata keys in the h5mu .obs to include in the output JSON.",
+          "example" : [
+            "donor_id;cell_type;batch;condition"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for CellRanger reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_cellbender",
+          "description" : "The cellbender keys in the h5mu .obs to include in the output JSON",
+          "default" : [
+            "cellbender_background_fraction",
+            "cellbender_cell_probability",
+            "cellbender_cell_size",
+            "cellbender_droplet_efficiency"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_cellranger_metrics",
+          "description" : "The key in the h5mu file .uns that contains the cellranger metrics",
+          "default" : [
+            "metrics_cellranger"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for Xenium reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_nucleus_area",
+          "description" : "The key in the h5mu .obs field that contains the nucleus area.",
+          "default" : [
+            "nucleus_area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_area",
+          "description" : "The key in the h5mu .obs field that contains the cell area.",
+          "default" : [
+            "cell_area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_x_coord",
+          "description" : "The key in the h5mu .obs field that contains the x coordinate.",
+          "default" : [
+            "x_coord"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_y_coord",
+          "description" : "The key in the h5mu .obs field that contains the y coordinate.",
+          "default" : [
+            "y_coord"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_control_probe_counts",
+          "description" : "The key in the h5mu .obs field that contains the number of control probes.",
+          "default" : [
+            "control_probe_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_control_codeword_counts",
+          "description" : "The key in the h5mu .obs field that contains the number of control codewords.",
+          "default" : [
+            "control_codeword_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "report_structure"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and processed by the QC workflow, and generates:\n- A JSON file that contains the combined data for the QC report\n- A JSON file that defines the layout and structure of the QC report\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "repositories" : [
+    {
+      "type" : "github",
+      "name" : "openpipeline",
+      "repo" : "openpipelines-bio/openpipeline",
+      "tag" : "2.1.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "build_main",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json",
+    "viash_version" : "0.9.4",
+    "git_commit" : "dfd68e55fafa27fc489f14bfab8ad04315f7f233",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_incubator",
+    "version" : "build_main",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "github",
+        "name" : "openpipeline",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'build_main'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import json
+import pandas as pd
+from pathlib import Path
+import anndata as ad
+import h5py
+import sys
+import os
+import shutil
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'ingestion_method': $( if [ ! -z ${VIASH_PAR_INGESTION_METHOD+x} ]; then echo "r'${VIASH_PAR_INGESTION_METHOD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_sample_id': $( if [ ! -z ${VIASH_PAR_OBS_SAMPLE_ID+x} ]; then echo "r'${VIASH_PAR_OBS_SAMPLE_ID//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_total_counts': $( if [ ! -z ${VIASH_PAR_OBS_TOTAL_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_TOTAL_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OBS_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_fraction_mitochondrial': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_fraction_ribosomal': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_RIBOSOMAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_RIBOSOMAL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_reporting_json': $( if [ ! -z ${VIASH_PAR_OUTPUT_REPORTING_JSON+x} ]; then echo "r'${VIASH_PAR_OUTPUT_REPORTING_JSON//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'min_total_counts': $( if [ ! -z ${VIASH_PAR_MIN_TOTAL_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MIN_TOTAL_COUNTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_MIN_NUM_NONZERO_VARS+x} ]; then echo "int(r'${VIASH_PAR_MIN_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'obs_metadata': $( if [ ! -z ${VIASH_PAR_OBS_METADATA+x} ]; then echo "r'${VIASH_PAR_OBS_METADATA//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_cellbender': $( if [ ! -z ${VIASH_PAR_OBS_CELLBENDER+x} ]; then echo "r'${VIASH_PAR_OBS_CELLBENDER//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'uns_cellranger_metrics': $( if [ ! -z ${VIASH_PAR_UNS_CELLRANGER_METRICS+x} ]; then echo "r'${VIASH_PAR_UNS_CELLRANGER_METRICS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_nucleus_area': $( if [ ! -z ${VIASH_PAR_OBS_NUCLEUS_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_NUCLEUS_AREA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_area': $( if [ ! -z ${VIASH_PAR_OBS_CELL_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_AREA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_x_coord': $( if [ ! -z ${VIASH_PAR_OBS_X_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_X_COORD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_y_coord': $( if [ ! -z ${VIASH_PAR_OBS_Y_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_Y_COORD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_control_probe_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_control_codeword_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+par["obs_cellbender"] = {} if not par["obs_cellbender"] else par["obs_cellbender"]
+
+
+def transform_df(df):
+    """Transform a DataFrame into the annotation object format."""
+    columns = []
+    for name in df.columns:
+        data = df[name]
+
+        # Determine dtype
+        if pd.api.types.is_integer_dtype(data):
+            dtype = "integer"
+        elif pd.api.types.is_float_dtype(data):
+            dtype = "numeric"
+        elif pd.api.types.is_categorical_dtype(data):
+            dtype = "categorical"
+        else:
+            raise ValueError(f"Unknown/unsupported data type for column {name}")
+
+        column_info = {"name": name, "dtype": dtype}
+
+        if dtype == "categorical":
+            column_info["data"] = data.cat.codes.tolist()
+            column_info["categories"] = data.cat.categories.tolist()
+        else:
+            column_info["data"] = [None if pd.isna(x) else x for x in data]
+
+        columns.append(column_info)
+
+    return {"num_rows": len(df), "num_cols": len(df.columns), "columns": columns}
+
+
+def check_optional_obs_keys(obs, keys, message):
+    missing_keys = [key for key in keys if key not in obs.columns]
+    if missing_keys:
+        logger.info(f"Missing keys in obs: {', '.join(missing_keys)}. {message}")
+
+
+def transform_cellranger_metrics(uns, sample_id):
+    if not par["uns_cellranger_metrics"] in uns:
+        raise ValueError(f"Could not find cellranger metrics in uns: {par['uns_cellranger_metrics']}. Provide correct value for --uns_cellranger_metrics or make sure data was ingested using CellRanger multi.")
+
+    cellranger_metrics = (
+        uns[par["uns_cellranger_metrics"]]
+        .pivot_table(
+            index=[],
+            columns="Metric Name",
+            values="Metric Value",
+            aggfunc="first",
+        )
+        .reset_index(drop=True)
+    )
+
+    cellranger_metrics.columns.name = None
+    # Remove thousands separator and convert to numeric
+    cellranger_metrics = cellranger_metrics.map(
+        lambda x: (
+            pd.to_numeric(x.replace(",", ""), errors="coerce")
+            if isinstance(x, str)
+            else x
+        )
+    )
+    # Replace spaces with underscores in column names
+    cellranger_metrics.columns = cellranger_metrics.columns.str.replace(" ", "_")
+    for col in cellranger_metrics.columns:
+        cellranger_metrics[col] = pd.to_numeric(cellranger_metrics[col], errors="coerce")
+    cellranger_metrics["sample_id"] = [sample_id[0]]
+
+    return cellranger_metrics
+
+
+def format_cellbender_columns(mod_obs):
+    # Check if celbender was run on the dataset
+    if par["obs_cellbender"]:
+        check_optional_obs_keys(mod_obs, par["obs_cellbender"], "Run cellbender first to include these metrics.")
+
+    cellbender_obs_keys = [column for column in par["obs_cellbender"] if column in mod_obs]
+
+    for key in cellbender_obs_keys:
+        if not pd.api.types.is_float_dtype(mod_obs[key]):
+            try:
+                mod_obs[key] = mod_obs[key].astype("float16")
+            except ValueError:
+                raise ValueError(f"Could not convert column {key} to a float dtype. Please make sure all cellbender metrics are numeric.")
+
+    return cellbender_obs_keys, mod_obs
+
+
+def format_required_columns(required_keys, mod_obs):
+
+    for key in required_keys:
+        if not pd.api.types.is_numeric_dtype(mod_obs[key]):
+            raise ValueError(f"Column {key} must be a numeric dtype.")
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_total_counts"]]):
+        logger.info(f"Converting {par['obs_total_counts']} from {mod_obs[par['obs_total_counts']].dtype} to integer dtype...")
+        mod_obs[par["obs_total_counts"]] = mod_obs[par["obs_total_counts"]].astype(int)
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_num_nonzero_vars"]]):
+        logger.info(f"Converting {par['obs_num_nonzero_vars']} from {mod_obs[par['obs_num_nonzero_vars']].dtype} to integer dtype...")
+        mod_obs[par["obs_num_nonzero_vars"]] = mod_obs[par["obs_num_nonzero_vars"]].astype(int)
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_mitochondrial"]]):
+        logger.info(f"Converting {par['obs_fraction_mitochondrial']} from {mod_obs[par['obs_fraction_mitochondrial']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_mitochondrial"]] = mod_obs[par["obs_fraction_mitochondrial"]].astype("float16")
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_ribosomal"]]):
+        logger.info(f"Converting {par['obs_fraction_ribosomal']} from {mod_obs[par['obs_fraction_ribosomal']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_ribosomal"]] = mod_obs[par["obs_fraction_ribosomal"]].astype("float16")
+
+    return mod_obs
+
+
+def format_categorical_columns(mod_obs):
+    # Fetch all categorical columns for grouping if no columns are provided
+    if not par["obs_metadata"]:
+        metadata_obs_keys = mod_obs.select_dtypes(include=["object", "category"]).columns.tolist()
+        if par["obs_sample_id"] in metadata_obs_keys:
+            metadata_obs_keys.remove(par["obs_sample_id"])
+    else:
+        check_optional_obs_keys(mod_obs, par["obs_metadata"], "Make sure requested metadata colmuns are present in obs.")
+        metadata_obs_keys = [key for key in par["obs_metadata"] if key in mod_obs]
+
+    for key in metadata_obs_keys:
+        if not isinstance(key, pd.CategoricalDtype):
+            logger.info(f"{key} is not a categorical dtype. Converting {key} from {mod_obs[key].dtype} to categorical dtype...")
+            mod_obs[key] = mod_obs[key].astype(str).astype("category")
+
+    return metadata_obs_keys, mod_obs
+
+
+def generate_cellranger_stats(mod_obs, uns, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Fetch and format cellbender columns
+    cellbender_obs_keys, mod_obs = format_cellbender_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in cellbender_obs_keys},
+            **{key: mod_obs[key] for key in metadata_obs_keys},
+        }
+    )
+
+    cellranger_stats = transform_cellranger_metrics(uns, sample_id)
+
+    return cell_rna_stats, cellranger_stats
+
+
+def format_xenium_columns(mod_obs):
+
+    mod_obs["nucleus_ratio"] = mod_obs[par["obs_nucleus_area"]] / mod_obs[par["obs_cell_area"]]
+
+    xenium_formatted_columns = [par["obs_cell_area"], "nucleus_ratio", "x_coord", "y_coord"]
+    for key in xenium_formatted_columns:
+        mod_obs[key] = mod_obs[key].astype("float16")
+
+    return mod_obs, xenium_formatted_columns
+
+
+def generate_xenium_stats(mod_obs, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format xenium-specific columns
+    mod_obs, xenium_formatted_columns = format_xenium_columns(mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in xenium_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+
+    return cell_rna_stats
+
+
+def concatenate_dataframes(dfs):
+    \'\'\'Concatenates a list of dataframes into a single dataframe, preserving categorical columns.\'\'\'
+    df = pd.concat(dfs, ignore_index=True)
+
+    # Find categorical columns that became object columms
+    for col in df.columns:
+        if any(df[col].dtype.name == 'category' for df in dfs if col in df.columns):
+            # Get all categorical series for this column
+            cat_series = [df[col] for df in dfs if col in df.columns and df[col].dtype.name == 'category']
+            if cat_series:
+                # Union the categories and apply to result
+                unioned = pd.api.types.union_categoricals(cat_series)
+                df[col] = pd.Categorical(df[col], categories=unioned.categories)
+    return df
+
+
+def main(par):
+    cell_stats_dfs = []
+    sample_stats_dfs = []
+    metrics_cellranger_dfs = []
+
+    for i, mudata_file in enumerate(par["input"]):
+        logger.info(f"Processing {mudata_file}")
+
+        # read h5mu file
+        file = h5py.File(mudata_file, "r")
+
+        # read the necessary info
+        grp_mod = file["mod"][par["modality"]]
+        mod_obs = ad.experimental.read_elem(grp_mod["obs"])
+        mod_obsm = ad.experimental.read_elem(grp_mod["obsm"])
+        uns = ad.experimental.read_elem(file["uns"])
+
+        # close the h5mu file
+        file.close()
+
+        barcodes_original_count = mod_obs.shape[0]
+
+        # Add coordinates to obs before filtering
+        if par["ingestion_method"] == "xenium":
+            mod_obs["x_coord"] = mod_obsm["spatial"][:, 0]
+            mod_obs["y_coord"] = mod_obsm["spatial"][:, 1]
+
+        # Pre-filter cells
+        logger.info("Pre-filtering cells based on counts...")
+        if "min_total_counts" in par:
+            mod_obs = mod_obs[mod_obs["total_counts"] >= par["min_total_counts"]]
+        if "min_num_nonzero_vars" in par:
+            mod_obs = mod_obs[mod_obs["num_nonzero_vars"] >= par["min_num_nonzero_vars"]]
+        barcodes_filtered_count = mod_obs.shape[0]
+
+        # Detect sample id's
+        logger.info("Detecting sample id's...")
+        sample_id = (
+            mod_obs[par["obs_sample_id"]].tolist()
+            if par["obs_sample_id"] in mod_obs.columns
+            else [f"sample_{i}"] * mod_obs.shape[0]
+        )
+
+        # Generating sample summary statistics
+        logger.info("Generating sample summary statistics...")
+        required_keys = [
+            par["obs_total_counts"],
+            par["obs_num_nonzero_vars"],
+            par["obs_fraction_mitochondrial"],
+            par["obs_fraction_ribosomal"]
+            ]
+        missing_keys = [key for key in required_keys if key not in mod_obs.columns]
+        if missing_keys:
+            raise ValueError(f"Missing keys in obs: {', '.join(missing_keys)}")
+
+        sample_summary = {
+            "sample_id": pd.Categorical([sample_id[0]]),
+            "rna_num_barcodes": [barcodes_original_count],
+            "rna_num_barcodes_filtered": [barcodes_filtered_count],
+            "rna_sum_total_counts": [mod_obs[par["obs_total_counts"]].sum()],
+            "rna_median_total_counts": [mod_obs[par["obs_total_counts"]].median()],
+            "rna_overall_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].sum()],
+            "rna_median_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].median()],
+        }
+
+        if par["ingestion_method"] == "xenium":
+            sample_summary["control_probe_percentage"] = mod_obs[par["obs_control_probe_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+            sample_summary["negative_decoding_percentage"] = mod_obs[par["obs_control_codeword_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+
+        sample_summary_stats = pd.DataFrame(sample_summary)
+
+        if par["ingestion_method"] == "cellranger_multi":
+            cell_rna_stats, cellranger_stats = generate_cellranger_stats(mod_obs, uns, sample_id, required_keys)
+            metrics_cellranger_dfs.append(cellranger_stats)
+
+        if par["ingestion_method"] == "xenium":
+            cell_rna_stats = generate_xenium_stats(mod_obs, sample_id, required_keys)
+
+        cell_stats_dfs.append(cell_rna_stats)
+        sample_stats_dfs.append(sample_summary_stats)
+
+    # Combine dataframes of all samples
+    logger.info("Combining data of all samples into single object...")
+    combined_cell_stats = concatenate_dataframes(cell_stats_dfs)
+    combined_sample_stats = concatenate_dataframes(sample_stats_dfs)
+    if par["ingestion_method"] == "cellranger_multi":
+        combined_metrics_cellranger = concatenate_dataframes(metrics_cellranger_dfs)
+
+    report_categories = [combined_cell_stats, combined_sample_stats]
+
+    if par["ingestion_method"] == "cellranger_multi":
+        report_categories.append(combined_metrics_cellranger)
+
+    for df in report_categories:
+        df["sample_id"] = pd.Categorical(df["sample_id"])
+
+    output = {
+        "cell_rna_stats": transform_df(combined_cell_stats),
+        "sample_summary_stats": transform_df(combined_sample_stats)
+    }
+
+    if par["ingestion_method"] == "cellranger_multi":
+        output["metrics_cellranger_stats"] = transform_df(combined_metrics_cellranger)
+
+    logger.info(f"Writing output data json to {par['output']}")
+    output_path = Path(par["output"])
+    with open(output_path, "w") as f:
+        json.dump(output, f, indent=2)
+
+    report_structures = {
+        "cellranger_multi": os.path.join(meta["resources_dir"], "report_structure/cellranger.json"),
+        "xenium": os.path.join(meta["resources_dir"], "report_structure/xenium.json")
+    }
+
+    logger.info(f"Writing output report structure json to {par['output_reporting_json']}")
+    shutil.copy(report_structures[par["ingestion_method"]], par["output_reporting_json"])
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline_incubator/ingestion_qc/h5mu_to_qc_json",
+    "tag" : "build_main"
+  },
+  "label" : [
+    "midmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config
new file mode 100644
index 0000000..4cc5c93
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'ingestion_qc/h5mu_to_qc_json'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'build_main'
+  description = 'Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and processed by the QC workflow, and generates:\n- A JSON file that contains the combined data for the QC report\n- A JSON file that defines the layout and structure of the QC report\n'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
new file mode 100644
index 0000000..18769ba
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
@@ -0,0 +1,162 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "SampleQC",
+      "key": "metrics_cellranger_stats",
+      "additionalAxes": false,
+      "defaultFilters": [
+        {
+          "type": "bar",
+          "field": "Number_of_reads_in_the_library",
+          "label": "Number of reads per library",
+          "description": "Sequencing depth per sample. Higher values generally indicate more comprehensive cell profiling.",
+          "nBins": 10,
+          "groupBy": "sample_id",
+          "xAxisType": "linear",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Confidently_mapped_reads_in_cells",
+          "label": "Confidently mapped reads in cells",
+          "description": "Number of reads that were mapped unambiguously to the reference genome within cell-containing droplets.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Estimated_number_of_cells",
+          "label": "Estimated number of cells",
+          "description": "CellRanger's estimate of the number of cells per sample based on the UMI count distribution.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Sequencing_saturation",
+          "label": "Sequencing saturation",
+          "description": "Fraction of reads that are duplicates of existing UMIs. Higher values suggest deeper sequencing coverage.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        }
+      ]
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "pct_of_counts_in_top_50_vars",
+          "label": "Fraction UMI in top 50 genes per cell",
+          "description": "Proportion of RNA molecules from the 50 most-expressed genes in each cell.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_probability",
+          "label": "CellBender cell probability",
+          "description": "CellBender's statistical confidence (0-1) that a barcode represents a real cell, with higher values indicating stronger confidence.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_background_fraction",
+          "label": "CellBender background fraction",
+          "description": "Estimated percentage of each cell's RNA that comes from the ambient solution rather than the cell itself.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_size",
+          "label": "CellBender cell size",
+          "description": "CellBender's estimate of the true number of RNA molecules in each cell after removing ambient contamination. Reflects actual cell RNA content rather than raw UMI counts.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_droplet_efficiency",
+          "label": "CellBender droplet efficiency",
+          "description": "CellBender's estimate of how efficiently each droplet captured RNA molecules. Higher values indicate more reliable RNA sampling within individual droplets.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
new file mode 100644
index 0000000..9dc7a8b
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
@@ -0,0 +1,91 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "cell_area",
+          "label": "Segmented cell area",
+          "description": "Area of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "nucleus_ratio",
+          "label": "Nucleus Ratio",
+          "description": "Ratio of the nucleus area to the segmented cell area.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
\ No newline at end of file
diff --git a/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/.config.vsh.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/.config.vsh.yaml
new file mode 100644
index 0000000..84ed89b
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/.config.vsh.yaml
@@ -0,0 +1,647 @@
+name: "cellbender_remove_background"
+namespace: "correction"
+version: "2.1.2"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file. Data file on which to run tool. Data must be un-filtered:\
+      \ it should include empty droplets."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "List of modalities to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Full count matrix as an h5mu file, with background RNA removed.\
+      \ This file contains all the original droplet barcodes."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer_output"
+    description: "Output layer"
+    info: null
+    default:
+    - "cellbender_corrected"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_background_fraction"
+    info: null
+    default:
+    - "cellbender_background_fraction"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_probability"
+    info: null
+    default:
+    - "cellbender_cell_probability"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_size"
+    info: null
+    default:
+    - "cellbender_cell_size"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_droplet_efficiency"
+    info: null
+    default:
+    - "cellbender_droplet_efficiency"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_latent_scale"
+    info: null
+    default:
+    - "cellbender_latent_scale"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_ambient_expression"
+    info: null
+    default:
+    - "cellbender_ambient_expression"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_gene_expression_encoding"
+    info: null
+    default:
+    - "cellbender_gene_expression_encoding"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Arguments"
+  arguments:
+  - type: "boolean"
+    name: "--expected_cells_from_qc"
+    description: "Will use the Cell Ranger QC to determine the estimated number of\
+      \ cells"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--expected_cells"
+    description: "Number of cells expected in the dataset (a rough estimate within\
+      \ a factor of 2 is sufficient)."
+    info: null
+    example:
+    - 1000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--total_droplets_included"
+    description: "The number of droplets from the rank-ordered UMI plot\nthat will\
+      \ have their cell probabilities inferred as an\noutput. Include the droplets\
+      \ which might contain cells.\nDroplets beyond TOTAL_DROPLETS_INCLUDED should\
+      \ be\n'surely empty' droplets.\n"
+    info: null
+    example:
+    - 25000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--force_cell_umi_prior"
+    description: "Ignore CellBender's heuristic prior estimation, and use this prior\
+      \ for UMI counts in cells."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--force_empty_umi_prior"
+    description: "Ignore CellBender's heuristic prior estimation, and use this prior\
+      \ for UMI counts in empty droplets."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--model"
+    description: "Which model is being used for count data.\n\n* 'naive' subtracts\
+      \ the estimated ambient profile.\n* 'simple' does not model either ambient RNA\
+      \ or random barcode swapping (for debugging purposes -- not recommended).\n\
+      * 'ambient' assumes background RNA is incorporated into droplets.\n* 'swapping'\
+      \ assumes background RNA comes from random barcode swapping (via PCR chimeras).\n\
+      * 'full' uses a combined ambient and swapping model.\n"
+    info: null
+    default:
+    - "full"
+    required: false
+    choices:
+    - "naive"
+    - "simple"
+    - "ambient"
+    - "swapping"
+    - "full"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--epochs"
+    description: "Number of epochs to train."
+    info: null
+    default:
+    - 150
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--low_count_threshold"
+    description: "Droplets with UMI counts below this number are completely \nexcluded\
+      \ from the analysis. This can help identify the correct \nprior for empty droplet\
+      \ counts in the rare case where empty \ncounts are extremely high (over 200).\n"
+    info: null
+    default:
+    - 5
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--z_dim"
+    description: "Dimension of latent variable z.\n"
+    info: null
+    default:
+    - 64
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--z_layers"
+    description: "Dimension of hidden layers in the encoder for z.\n"
+    info: null
+    default:
+    - 512
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--training_fraction"
+    description: "Training detail: the fraction of the data used for training.\nThe\
+      \ rest is never seen by the inference algorithm. Speeds up learning.\n"
+    info: null
+    default:
+    - 0.9
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--empty_drop_training_fraction"
+    description: "Training detail: the fraction of the training data each epoch that\
+      \ \nis drawn (randomly sampled) from surely empty droplets.\n"
+    info: null
+    default:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--ignore_features"
+    description: "Integer indices of features to ignore entirely. In the output\n\
+      count matrix, the counts for these features will be unchanged.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--fpr"
+    description: "Target 'delta' false positive rate in [0, 1). Use 0 for a cohort\n\
+      of samples which will be jointly analyzed for differential expression.\nA false\
+      \ positive is a true signal count that is erroneously removed.\nMore background\
+      \ removal is accompanied by more signal removal at\nhigh values of FPR. You\
+      \ can specify multiple values, which will\ncreate multiple output files.\n"
+    info: null
+    default:
+    - 0.01
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--exclude_feature_types"
+    description: "Feature types to ignore during the analysis. These features will\n\
+      be left unchanged in the output file.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--projected_ambient_count_threshold"
+    description: "Controls how many features are included in the analysis, which\n\
+      can lead to a large speedup. If a feature is expected to have less\nthan PROJECTED_AMBIENT_COUNT_THRESHOLD\
+      \ counts total in all cells\n(summed), then that gene is excluded, and it will\
+      \ be unchanged\nin the output count matrix. For example, \nPROJECTED_AMBIENT_COUNT_THRESHOLD\
+      \ = 0 will include all features\nwhich have even a single count in any empty\
+      \ droplet.\n"
+    info: null
+    default:
+    - 0.1
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--learning_rate"
+    description: "Training detail: lower learning rate for inference.\nA OneCycle\
+      \ learning rate schedule is used, where the\nupper learning rate is ten times\
+      \ this value. (For this\nvalue, probably do not exceed 1e-3).\n"
+    info: null
+    default:
+    - 1.0E-4
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--final_elbo_fail_fraction"
+    description: "Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO\
+      \ - initial_test_ELBO) > FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically\
+      \ re-run if --num-training-tries > 1.\nBy default, will not fail training based\
+      \ on final_training_ELBO.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--epoch_elbo_fail_fraction"
+    description: "Training is considered to have failed if \n(previous_epoch_test_ELBO\
+      \ - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO)\
+      \ > EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries\
+      \ > 1.\nBy default, will not fail training based on epoch_training_ELBO.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_training_tries"
+    description: "Number of times to attempt to train the model. At each subsequent\
+      \ attempt,\nthe learning rate is multiplied by LEARNING_RATE_RETRY_MULT.\n"
+    info: null
+    default:
+    - 1
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--learning_rate_retry_mult"
+    description: "Learning rate is multiplied by this amount each time a new training\n\
+      attempt is made. (This parameter is only used if training fails based\non EPOCH_ELBO_FAIL_FRACTION\
+      \ or FINAL_ELBO_FAIL_FRACTION and\nNUM_TRAINING_TRIES is > 1.) \n"
+    info: null
+    default:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--posterior_batch_size"
+    description: "Training detail: size of batches when creating the posterior.\n\
+      Reduce this to avoid running out of GPU memory creating the posterior\n(will\
+      \ be slower).\n"
+    info: null
+    default:
+    - 128
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--posterior_regulation"
+    description: "Posterior regularization method. (For experts: not required for\
+      \ normal usage,\nsee documentation). \n\n* PRq is approximate quantile-targeting.\n\
+      * PRmu is approximate mean-targeting aggregated over genes (behavior of v0.2.0).\n\
+      * PRmu_gene is approximate mean-targeting per gene.\n"
+    info: null
+    required: false
+    choices:
+    - "PRq"
+    - "PRmu"
+    - "PRmu_gene"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--alpha"
+    description: "Tunable parameter alpha for the PRq posterior regularization method\n\
+      (not normally used: see documentation).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--q"
+    description: "Tunable parameter q for the CDF threshold estimation method (not\n\
+      normally used: see documentation).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--estimator"
+    description: "Output denoised count estimation method. (For experts: not required\n\
+      for normal usage, see documentation).\n"
+    info: null
+    default:
+    - "mckp"
+    required: false
+    choices:
+    - "map"
+    - "mean"
+    - "cdf"
+    - "sample"
+    - "mckp"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--estimator_multiple_cpu"
+    description: "Including the flag --estimator-multiple-cpu will use more than one\n\
+      CPU to compute the MCKP output count estimator in parallel (does nothing\nfor\
+      \ other estimators).\n"
+    info: null
+    direction: "input"
+  - type: "boolean"
+    name: "--constant_learning_rate"
+    description: "Including the flag --constant-learning-rate will use the ClippedAdam\n\
+      optimizer instead of the OneCycleLR learning rate schedule, which is\nthe default.\
+      \ Learning is faster with the OneCycleLR schedule.\nHowever, training can easily\
+      \ be continued from a checkpoint for more\nepochs than the initial command specified\
+      \ when using ClippedAdam. On\nthe other hand, if using the OneCycleLR schedule\
+      \ with 150 epochs\nspecified, it is not possible to pick up from that final\
+      \ checkpoint\nand continue training until 250 epochs.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--debug"
+    description: "Including the flag --debug will log extra messages useful for debugging.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--cuda"
+    description: "Including the flag --cuda will run the inference on a\nGPU.\n"
+    info: null
+    direction: "input"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Eliminating technical artifacts from high-throughput single-cell RNA\
+  \ sequencing data.\n\nThis module removes counts due to ambient RNA molecules and\
+  \ random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the\
+  \ moment, only the count matrices produced by the CellRanger count pipeline is supported.\
+  \ Support for additional tools and protocols \nwill be added in the future. A quick\
+  \ start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midcpu"
+    - "midmem"
+    - "gpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "nvcr.io/nvidia/cuda:11.8.0-devel-ubuntu22.04"
+  target_tag: "2.1.0"
+  namespace_separator: "/"
+  setup:
+  - type: "docker"
+    run:
+    - "apt update && DEBIAN_FRONTEND=noninteractive apt install -y make build-essential\
+      \ libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget ca-certificates\
+      \ curl llvm libncurses5-dev xz-utils tk-dev libxml2-dev libxmlsec1-dev libffi-dev\
+      \ liblzma-dev mecab-ipadic-utf8 git \\\n&& curl https://pyenv.run | bash \\\n\
+      && pyenv update \\\n&& pyenv install $PYTHON_VERSION \\\n&& pyenv global $PYTHON_VERSION\
+      \ \\\n&& apt-get clean\n"
+    env:
+    - "PYENV_ROOT=\"/root/.pyenv\""
+    - "PATH=\"$PYENV_ROOT/shims:$PYENV_ROOT/bin:$PATH\""
+    - "PYTHON_VERSION=3.7.16"
+  - type: "python"
+    user: false
+    packages:
+    - "lxml~=4.8.0"
+    - "mudata~=0.2.1"
+    - "cellbender~=0.3.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+build_info:
+  config: "src/correction/cellbender_remove_background/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker"
+  output: "target/nextflow/correction/cellbender_remove_background"
+  executable: "target/nextflow/correction/cellbender_remove_background/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "a0c9522486585774f76416150f8a3291409b5363"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  git_tag: "2.1.1-2-ga0c95224865"
+package_config:
+  name: "openpipeline"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".version := \"2.1.2\""
+  - ".engines[.type == 'docker'].target_tag := '2.1.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "openpipelines-bio"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/main.nf b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/main.nf
new file mode 100644
index 0000000..3c79f75
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/main.nf
@@ -0,0 +1,4517 @@
+// cellbender_remove_background 2.1.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "cellbender_remove_background",
+  "namespace" : "correction",
+  "version" : "2.1.2",
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file. Data file on which to run tool. Data must be un-filtered: it should include empty droplets.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "List of modalities to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Full count matrix as an h5mu file, with background RNA removed. This file contains all the original droplet barcodes.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer_output",
+          "description" : "Output layer",
+          "default" : [
+            "cellbender_corrected"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_background_fraction",
+          "default" : [
+            "cellbender_background_fraction"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_probability",
+          "default" : [
+            "cellbender_cell_probability"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_size",
+          "default" : [
+            "cellbender_cell_size"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_droplet_efficiency",
+          "default" : [
+            "cellbender_droplet_efficiency"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_latent_scale",
+          "default" : [
+            "cellbender_latent_scale"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_ambient_expression",
+          "default" : [
+            "cellbender_ambient_expression"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_gene_expression_encoding",
+          "default" : [
+            "cellbender_gene_expression_encoding"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--expected_cells_from_qc",
+          "description" : "Will use the Cell Ranger QC to determine the estimated number of cells",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--expected_cells",
+          "description" : "Number of cells expected in the dataset (a rough estimate within a factor of 2 is sufficient).",
+          "example" : [
+            1000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--total_droplets_included",
+          "description" : "The number of droplets from the rank-ordered UMI plot\nthat will have their cell probabilities inferred as an\noutput. Include the droplets which might contain cells.\nDroplets beyond TOTAL_DROPLETS_INCLUDED should be\n'surely empty' droplets.\n",
+          "example" : [
+            25000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--force_cell_umi_prior",
+          "description" : "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in cells.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--force_empty_umi_prior",
+          "description" : "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in empty droplets.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--model",
+          "description" : "Which model is being used for count data.\n\n* 'naive' subtracts the estimated ambient profile.\n* 'simple' does not model either ambient RNA or random barcode swapping (for debugging purposes -- not recommended).\n* 'ambient' assumes background RNA is incorporated into droplets.\n* 'swapping' assumes background RNA comes from random barcode swapping (via PCR chimeras).\n* 'full' uses a combined ambient and swapping model.\n",
+          "default" : [
+            "full"
+          ],
+          "required" : false,
+          "choices" : [
+            "naive",
+            "simple",
+            "ambient",
+            "swapping",
+            "full"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--epochs",
+          "description" : "Number of epochs to train.",
+          "default" : [
+            150
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--low_count_threshold",
+          "description" : "Droplets with UMI counts below this number are completely \nexcluded from the analysis. This can help identify the correct \nprior for empty droplet counts in the rare case where empty \ncounts are extremely high (over 200).\n",
+          "default" : [
+            5
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--z_dim",
+          "description" : "Dimension of latent variable z.\n",
+          "default" : [
+            64
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--z_layers",
+          "description" : "Dimension of hidden layers in the encoder for z.\n",
+          "default" : [
+            512
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--training_fraction",
+          "description" : "Training detail: the fraction of the data used for training.\nThe rest is never seen by the inference algorithm. Speeds up learning.\n",
+          "default" : [
+            0.9
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--empty_drop_training_fraction",
+          "description" : "Training detail: the fraction of the training data each epoch that \nis drawn (randomly sampled) from surely empty droplets.\n",
+          "default" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--ignore_features",
+          "description" : "Integer indices of features to ignore entirely. In the output\ncount matrix, the counts for these features will be unchanged.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--fpr",
+          "description" : "Target 'delta' false positive rate in [0, 1). Use 0 for a cohort\nof samples which will be jointly analyzed for differential expression.\nA false positive is a true signal count that is erroneously removed.\nMore background removal is accompanied by more signal removal at\nhigh values of FPR. You can specify multiple values, which will\ncreate multiple output files.\n",
+          "default" : [
+            0.01
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--exclude_feature_types",
+          "description" : "Feature types to ignore during the analysis. These features will\nbe left unchanged in the output file.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--projected_ambient_count_threshold",
+          "description" : "Controls how many features are included in the analysis, which\ncan lead to a large speedup. If a feature is expected to have less\nthan PROJECTED_AMBIENT_COUNT_THRESHOLD counts total in all cells\n(summed), then that gene is excluded, and it will be unchanged\nin the output count matrix. For example, \nPROJECTED_AMBIENT_COUNT_THRESHOLD = 0 will include all features\nwhich have even a single count in any empty droplet.\n",
+          "default" : [
+            0.1
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--learning_rate",
+          "description" : "Training detail: lower learning rate for inference.\nA OneCycle learning rate schedule is used, where the\nupper learning rate is ten times this value. (For this\nvalue, probably do not exceed 1e-3).\n",
+          "default" : [
+            1.0E-4
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--final_elbo_fail_fraction",
+          "description" : "Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO - initial_test_ELBO) > FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on final_training_ELBO.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--epoch_elbo_fail_fraction",
+          "description" : "Training is considered to have failed if \n(previous_epoch_test_ELBO - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO) > EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on epoch_training_ELBO.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_training_tries",
+          "description" : "Number of times to attempt to train the model. At each subsequent attempt,\nthe learning rate is multiplied by LEARNING_RATE_RETRY_MULT.\n",
+          "default" : [
+            1
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--learning_rate_retry_mult",
+          "description" : "Learning rate is multiplied by this amount each time a new training\nattempt is made. (This parameter is only used if training fails based\non EPOCH_ELBO_FAIL_FRACTION or FINAL_ELBO_FAIL_FRACTION and\nNUM_TRAINING_TRIES is > 1.) \n",
+          "default" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--posterior_batch_size",
+          "description" : "Training detail: size of batches when creating the posterior.\nReduce this to avoid running out of GPU memory creating the posterior\n(will be slower).\n",
+          "default" : [
+            128
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--posterior_regulation",
+          "description" : "Posterior regularization method. (For experts: not required for normal usage,\nsee documentation). \n\n* PRq is approximate quantile-targeting.\n* PRmu is approximate mean-targeting aggregated over genes (behavior of v0.2.0).\n* PRmu_gene is approximate mean-targeting per gene.\n",
+          "required" : false,
+          "choices" : [
+            "PRq",
+            "PRmu",
+            "PRmu_gene"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--alpha",
+          "description" : "Tunable parameter alpha for the PRq posterior regularization method\n(not normally used: see documentation).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--q",
+          "description" : "Tunable parameter q for the CDF threshold estimation method (not\nnormally used: see documentation).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--estimator",
+          "description" : "Output denoised count estimation method. (For experts: not required\nfor normal usage, see documentation).\n",
+          "default" : [
+            "mckp"
+          ],
+          "required" : false,
+          "choices" : [
+            "map",
+            "mean",
+            "cdf",
+            "sample",
+            "mckp"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--estimator_multiple_cpu",
+          "description" : "Including the flag --estimator-multiple-cpu will use more than one\nCPU to compute the MCKP output count estimator in parallel (does nothing\nfor other estimators).\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--constant_learning_rate",
+          "description" : "Including the flag --constant-learning-rate will use the ClippedAdam\noptimizer instead of the OneCycleLR learning rate schedule, which is\nthe default. Learning is faster with the OneCycleLR schedule.\nHowever, training can easily be continued from a checkpoint for more\nepochs than the initial command specified when using ClippedAdam. On\nthe other hand, if using the OneCycleLR schedule with 150 epochs\nspecified, it is not possible to pick up from that final checkpoint\nand continue training until 250 epochs.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--debug",
+          "description" : "Including the flag --debug will log extra messages useful for debugging.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--cuda",
+          "description" : "Including the flag --cuda will run the inference on a\nGPU.\n",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midcpu",
+          "midmem",
+          "gpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "nvcr.io/nvidia/cuda:11.8.0-devel-ubuntu22.04",
+      "target_tag" : "2.1.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "docker",
+          "run" : [
+            "apt update && DEBIAN_FRONTEND=noninteractive apt install -y make build-essential libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget ca-certificates curl llvm libncurses5-dev xz-utils tk-dev libxml2-dev libxmlsec1-dev libffi-dev liblzma-dev mecab-ipadic-utf8 git \\\\\n&& curl https://pyenv.run | bash \\\\\n&& pyenv update \\\\\n&& pyenv install $PYTHON_VERSION \\\\\n&& pyenv global $PYTHON_VERSION \\\\\n&& apt-get clean\n"
+          ],
+          "env" : [
+            "PYENV_ROOT=\\"/root/.pyenv\\"",
+            "PATH=\\"$PYENV_ROOT/shims:$PYENV_ROOT/bin:$PATH\\"",
+            "PYTHON_VERSION=3.7.16"
+          ]
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "lxml~=4.8.0",
+            "mudata~=0.2.1",
+            "cellbender~=0.3.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    }
+  ],
+  "build_info" : {
+    "config" : "/home/runner/work/openpipeline/openpipeline/src/correction/cellbender_remove_background/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker",
+    "output" : "/home/runner/work/openpipeline/openpipeline/target/nextflow/correction/cellbender_remove_background",
+    "viash_version" : "0.9.4",
+    "git_commit" : "a0c9522486585774f76416150f8a3291409b5363",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline",
+    "git_tag" : "2.1.1-2-ga0c95224865"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/home/runner/work/openpipeline/openpipeline/src",
+    "target" : "/home/runner/work/openpipeline/openpipeline/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".version := \\"2.1.2\\"",
+      ".engines[.type == 'docker'].target_tag := '2.1.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "openpipelines-bio",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata as mu
+import tempfile
+import subprocess
+import os
+import sys
+import numpy as np
+from scipy.sparse import csr_matrix
+from cellbender.remove_background.downstream import anndata_from_h5
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer_output': $( if [ ! -z ${VIASH_PAR_LAYER_OUTPUT+x} ]; then echo "r'${VIASH_PAR_LAYER_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_background_fraction': $( if [ ! -z ${VIASH_PAR_OBS_BACKGROUND_FRACTION+x} ]; then echo "r'${VIASH_PAR_OBS_BACKGROUND_FRACTION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_probability': $( if [ ! -z ${VIASH_PAR_OBS_CELL_PROBABILITY+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_PROBABILITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_size': $( if [ ! -z ${VIASH_PAR_OBS_CELL_SIZE+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_SIZE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_droplet_efficiency': $( if [ ! -z ${VIASH_PAR_OBS_DROPLET_EFFICIENCY+x} ]; then echo "r'${VIASH_PAR_OBS_DROPLET_EFFICIENCY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_latent_scale': $( if [ ! -z ${VIASH_PAR_OBS_LATENT_SCALE+x} ]; then echo "r'${VIASH_PAR_OBS_LATENT_SCALE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_ambient_expression': $( if [ ! -z ${VIASH_PAR_VAR_AMBIENT_EXPRESSION+x} ]; then echo "r'${VIASH_PAR_VAR_AMBIENT_EXPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_gene_expression_encoding': $( if [ ! -z ${VIASH_PAR_OBSM_GENE_EXPRESSION_ENCODING+x} ]; then echo "r'${VIASH_PAR_OBSM_GENE_EXPRESSION_ENCODING//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'expected_cells_from_qc': $( if [ ! -z ${VIASH_PAR_EXPECTED_CELLS_FROM_QC+x} ]; then echo "r'${VIASH_PAR_EXPECTED_CELLS_FROM_QC//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'expected_cells': $( if [ ! -z ${VIASH_PAR_EXPECTED_CELLS+x} ]; then echo "int(r'${VIASH_PAR_EXPECTED_CELLS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'total_droplets_included': $( if [ ! -z ${VIASH_PAR_TOTAL_DROPLETS_INCLUDED+x} ]; then echo "int(r'${VIASH_PAR_TOTAL_DROPLETS_INCLUDED//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'force_cell_umi_prior': $( if [ ! -z ${VIASH_PAR_FORCE_CELL_UMI_PRIOR+x} ]; then echo "int(r'${VIASH_PAR_FORCE_CELL_UMI_PRIOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'force_empty_umi_prior': $( if [ ! -z ${VIASH_PAR_FORCE_EMPTY_UMI_PRIOR+x} ]; then echo "int(r'${VIASH_PAR_FORCE_EMPTY_UMI_PRIOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'model': $( if [ ! -z ${VIASH_PAR_MODEL+x} ]; then echo "r'${VIASH_PAR_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'epochs': $( if [ ! -z ${VIASH_PAR_EPOCHS+x} ]; then echo "int(r'${VIASH_PAR_EPOCHS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'low_count_threshold': $( if [ ! -z ${VIASH_PAR_LOW_COUNT_THRESHOLD+x} ]; then echo "int(r'${VIASH_PAR_LOW_COUNT_THRESHOLD//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'z_dim': $( if [ ! -z ${VIASH_PAR_Z_DIM+x} ]; then echo "int(r'${VIASH_PAR_Z_DIM//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'z_layers': $( if [ ! -z ${VIASH_PAR_Z_LAYERS+x} ]; then echo "list(map(int, r'${VIASH_PAR_Z_LAYERS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'training_fraction': $( if [ ! -z ${VIASH_PAR_TRAINING_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_TRAINING_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'empty_drop_training_fraction': $( if [ ! -z ${VIASH_PAR_EMPTY_DROP_TRAINING_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_EMPTY_DROP_TRAINING_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'ignore_features': $( if [ ! -z ${VIASH_PAR_IGNORE_FEATURES+x} ]; then echo "list(map(int, r'${VIASH_PAR_IGNORE_FEATURES//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'fpr': $( if [ ! -z ${VIASH_PAR_FPR+x} ]; then echo "list(map(float, r'${VIASH_PAR_FPR//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'exclude_feature_types': $( if [ ! -z ${VIASH_PAR_EXCLUDE_FEATURE_TYPES+x} ]; then echo "r'${VIASH_PAR_EXCLUDE_FEATURE_TYPES//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'projected_ambient_count_threshold': $( if [ ! -z ${VIASH_PAR_PROJECTED_AMBIENT_COUNT_THRESHOLD+x} ]; then echo "float(r'${VIASH_PAR_PROJECTED_AMBIENT_COUNT_THRESHOLD//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'learning_rate': $( if [ ! -z ${VIASH_PAR_LEARNING_RATE+x} ]; then echo "float(r'${VIASH_PAR_LEARNING_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'final_elbo_fail_fraction': $( if [ ! -z ${VIASH_PAR_FINAL_ELBO_FAIL_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_FINAL_ELBO_FAIL_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'epoch_elbo_fail_fraction': $( if [ ! -z ${VIASH_PAR_EPOCH_ELBO_FAIL_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_EPOCH_ELBO_FAIL_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'num_training_tries': $( if [ ! -z ${VIASH_PAR_NUM_TRAINING_TRIES+x} ]; then echo "int(r'${VIASH_PAR_NUM_TRAINING_TRIES//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'learning_rate_retry_mult': $( if [ ! -z ${VIASH_PAR_LEARNING_RATE_RETRY_MULT+x} ]; then echo "float(r'${VIASH_PAR_LEARNING_RATE_RETRY_MULT//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'posterior_batch_size': $( if [ ! -z ${VIASH_PAR_POSTERIOR_BATCH_SIZE+x} ]; then echo "int(r'${VIASH_PAR_POSTERIOR_BATCH_SIZE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'posterior_regulation': $( if [ ! -z ${VIASH_PAR_POSTERIOR_REGULATION+x} ]; then echo "r'${VIASH_PAR_POSTERIOR_REGULATION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'alpha': $( if [ ! -z ${VIASH_PAR_ALPHA+x} ]; then echo "float(r'${VIASH_PAR_ALPHA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'q': $( if [ ! -z ${VIASH_PAR_Q+x} ]; then echo "float(r'${VIASH_PAR_Q//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'estimator': $( if [ ! -z ${VIASH_PAR_ESTIMATOR+x} ]; then echo "r'${VIASH_PAR_ESTIMATOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'estimator_multiple_cpu': $( if [ ! -z ${VIASH_PAR_ESTIMATOR_MULTIPLE_CPU+x} ]; then echo "r'${VIASH_PAR_ESTIMATOR_MULTIPLE_CPU//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'constant_learning_rate': $( if [ ! -z ${VIASH_PAR_CONSTANT_LEARNING_RATE+x} ]; then echo "r'${VIASH_PAR_CONSTANT_LEARNING_RATE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'debug': $( if [ ! -z ${VIASH_PAR_DEBUG+x} ]; then echo "r'${VIASH_PAR_DEBUG//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'cuda': $( if [ ! -z ${VIASH_PAR_CUDA+x} ]; then echo "r'${VIASH_PAR_CUDA//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+logger.info("Reading input mudata")
+mdata = mu.read_h5mu(par["input"])
+
+mod = par["modality"]
+logger.info("Performing log transformation on modality %s", mod)
+data = mdata.mod[mod]
+
+# import pathlib
+# with pathlib.Path(os.path.dirname(par["output"])) / "cellbender" as temp_dir:
+#     os.mkdir(temp_dir)
+with tempfile.TemporaryDirectory(
+    prefix="cellbender-", dir=meta["temp_dir"]
+) as temp_dir:
+    # construct paths within tempdir
+    input_file = os.path.join(temp_dir, "input.h5ad")
+    output_file = os.path.join(temp_dir, "output.h5")
+
+    logger.info("Creating AnnData input file for CellBender: '%s'", input_file)
+    data.write_h5ad(input_file)
+
+    logger.info("Constructing CellBender command")
+    cmd_pars = [
+        "cellbender",
+        "remove-background",
+        "--input",
+        input_file,
+        "--output",
+        output_file,
+        # don't create checkpoints because they're not used / returned anyways
+        "--checkpoint-mins",
+        "99999999",
+    ]
+
+    if meta.get("cpus") is not None:
+        cmd_pars += ["--cpu-threads", str(meta["cpus"])]
+
+    extra_args = [
+        ("--expected-cells", "expected_cells", True),
+        ("--total-droplets-included", "total_droplets_included", True),
+        ("--force-cell-umi-prior", "force_cell_umi_prior", True),
+        ("--force-empty-umi-prior", "force_empty_umi_prior", True),
+        ("--model", "model", True),
+        ("--epochs", "epochs", True),
+        ("--low-count-threshold", "low_count_threshold", True),
+        ("--z-dim", "z_dim", True),
+        ("--z-layers", "z_layers", True),
+        ("--training-fraction", "training_fraction", True),
+        ("--empty-drop-training-fraction", "empty_drop_training_fraction", True),
+        ("--ignore-features", "ignore_features", True),
+        ("--fpr", "fpr", True),
+        ("--exclude-feature-types", "exclude_feature_types", True),
+        (
+            "--projected-ambient-count-threshold",
+            "projected_ambient_count_threshold",
+            True,
+        ),
+        ("--learning-rate", "learning_rate", True),
+        ("--final-elbo-fail-fraction", "final_elbo_fail_fraction", True),
+        ("--epoch-elbo-fail-fraction", "epoch_elbo_fail_fraction", True),
+        ("--num-training-tries", "num_training_tries", True),
+        ("--learning-rate-retry-mult", "learning_rate_retry_mult", True),
+        ("--posterior-batch-size", "posterior_batch_size", True),
+        ("--posterior-regulation", "posterior_regulation", True),
+        ("--alpha", "alpha", True),
+        ("--q", "q", True),
+        ("--estimator", "estimator", True),
+        ("--estimator-multiple-cpu", "estimator_multiple_cpu", False),
+        ("--constant-learning-rate", "constant_learning_rate", False),
+        ("--debug", "debug", False),
+        ("--cuda", "cuda", False),
+    ]
+    for flag, name, is_kwarg in extra_args:
+        if par[name]:
+            values = par[name] if isinstance(par[name], list) else [par[name]]
+            cmd_pars += [flag] + [str(val) for val in values] if is_kwarg else [flag]
+
+    if par["expected_cells_from_qc"] and "metrics_cellranger" in data.uns:
+        assert (
+            par["expected_cells"] is None
+        ), "If min_counts is defined, expected_cells should be undefined"
+        assert (
+            par["total_droplets_included"] is None
+        ), "If min_counts is defined, expected_cells should be undefined"
+        met = data.uns["metrics_cellranger"]
+        col_name = "Estimated Number of Cells"
+        assert (
+            col_name in met.columns
+        ), "%s should be a column in .obs[metrics_cellranger]"
+        est_cells = met[col_name].values[0]
+        logger.info(
+            "Selecting --expected-cells %d and --total-droplets-included %d",
+            est_cells,
+            est_cells * 5,
+        )
+        cmd_pars += [
+            "--expected-cells",
+            str(est_cells),
+            "--total-droplets-included",
+            str(5 * est_cells),
+        ]
+
+    logger.info("Running CellBender: '%s'", " ".join(cmd_pars))
+    out = subprocess.check_output(cmd_pars).decode("utf-8")
+
+    logger.info("Reading CellBender 10xh5 output file: '%s'", output_file)
+    adata_out = anndata_from_h5(output_file, analyzed_barcodes_only=False)
+
+    logger.info("CellBender output format:", adata_out)
+
+    # AnnData object with n_obs x n_vars = 6794880 x 33538
+    #     obs: 'cellbender_analyzed'
+    #     var: 'ambient_expression', 'feature_type', 'genome', 'gene_id', 'cellbender_analyzed'
+    #     uns: 'background_fraction', 'barcode_indices_for_latents', 'cell_probability', 'cell_size', 'droplet_efficiency', 'gene_expression_encoding',
+    #          'cell_size_lognormal_std', 'empty_droplet_size_lognormal_loc', 'empty_droplet_size_lognormal_scale', 'swapping_fraction_dist_params',
+    #          'barcodes_analyzed', 'barcodes_analyzed_inds', 'estimator', 'features_analyzed_inds', 'fraction_data_used_for_testing', 'learning_curve_learning_rate_epoch',
+    #          'learning_curve_learning_rate_value', 'learning_curve_test_elbo', 'learning_curve_test_epoch', 'learning_curve_train_elbo', 'learning_curve_train_epoch',
+    #          'target_false_positive_rate'
+
+    logger.info("Copying X output to MuData")
+    data.layers[par["layer_output"]] = adata_out.X
+
+    logger.info("Copying .obs output to MuData")
+    obs_store = {
+        "obs_background_fraction": "background_fraction",
+        "obs_cell_probability": "cell_probability",
+        "obs_cell_size": "cell_size",
+        "obs_droplet_efficiency": "droplet_efficiency",
+        "obs_latent_scale": "latent_scale",
+    }
+    for to_name, from_name in obs_store.items():
+        if par[to_name]:
+            if from_name in adata_out.obs:
+                data.obs[par[to_name]] = adata_out.obs[from_name]
+            # when using unfiltered data, the values will be in uns instead of obs
+            elif (
+                from_name in adata_out.uns
+                and "barcode_indices_for_latents" in adata_out.uns
+            ):
+                vec = np.zeros(data.n_obs)
+                vec[adata_out.uns["barcode_indices_for_latents"]] = adata_out.uns[
+                    from_name
+                ]
+                data.obs[par[to_name]] = vec
+
+    logger.info("Copying .var output to MuData")
+    var_store = {"var_ambient_expression": "ambient_expression"}
+    for to_name, from_name in var_store.items():
+        if par[to_name]:
+            data.var[par[to_name]] = adata_out.var[from_name]
+
+    logger.info("Copying obsm_gene_expression_encoding output to MuData")
+    obsm_store = {"obsm_gene_expression_encoding": "gene_expression_encoding"}
+    for to_name, from_name in obsm_store.items():
+        if par[to_name]:
+            if from_name in adata_out.obsm:
+                data.obsm[par[to_name]] = adata_out.obsm[from_name]
+            elif (
+                from_name in adata_out.uns
+                and "barcode_indices_for_latents" in adata_out.uns
+            ):
+                matrix_to_store = adata_out.uns[from_name]
+                number_of_obs = data.X.shape[0]
+                latent_space_sparse = csr_matrix(
+                    (number_of_obs, par["z_dim"]), dtype=adata_out.uns[from_name].dtype
+                )
+                obs_rows_in_space_representation = adata_out.uns[
+                    "barcode_indices_for_latents"
+                ]
+                latent_space_sparse[obs_rows_in_space_representation] = adata_out.uns[
+                    from_name
+                ]
+                data.obsm[par[to_name]] = latent_space_sparse
+            else:
+                raise RuntimeError(
+                    "Requested to save latent gene encoding, but the data is either missing "
+                    "from cellbender output or in an incorrect format."
+                )
+
+
+logger.info("Writing to file %s", par["output"])
+mdata.write_h5mu(filename=par["output"], compression=par["output_compression"])
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "ghcr.io",
+    "image" : "openpipelines-bio/openpipeline/correction/cellbender_remove_background",
+    "tag" : "2.1.0"
+  },
+  "label" : [
+    "midcpu",
+    "midmem",
+    "gpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow.config
new file mode 100644
index 0000000..f6dc7cc
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow.config
@@ -0,0 +1,125 @@
+manifest {
+  name = 'correction/cellbender_remove_background'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = '2.1.2'
+  description = 'Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_params.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_params.yaml
new file mode 100644
index 0000000..cc465d7
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_params.yaml
@@ -0,0 +1,51 @@
+# Inputs
+input: # please fill in - example: "input.h5mu"
+modality: "rna"
+
+# Outputs
+# output: "$id.$key.output.h5mu"
+# output_compression: "gzip"
+layer_output: "cellbender_corrected"
+obs_background_fraction: "cellbender_background_fraction"
+obs_cell_probability: "cellbender_cell_probability"
+obs_cell_size: "cellbender_cell_size"
+obs_droplet_efficiency: "cellbender_droplet_efficiency"
+obs_latent_scale: "cellbender_latent_scale"
+var_ambient_expression: "cellbender_ambient_expression"
+obsm_gene_expression_encoding: "cellbender_gene_expression_encoding"
+
+# Arguments
+expected_cells_from_qc: false
+# expected_cells: 1000
+# total_droplets_included: 25000
+# force_cell_umi_prior: 123
+# force_empty_umi_prior: 123
+model: "full"
+epochs: 150
+low_count_threshold: 5
+z_dim: 64
+z_layers: [512]
+training_fraction: 0.9
+empty_drop_training_fraction: 0.2
+# ignore_features: [123]
+fpr: [0.01]
+# exclude_feature_types: ["foo"]
+projected_ambient_count_threshold: 0.1
+learning_rate: 1.0E-4
+# final_elbo_fail_fraction: 123.0
+# epoch_elbo_fail_fraction: 123.0
+num_training_tries: 1
+learning_rate_retry_mult: 0.2
+posterior_batch_size: 128
+# posterior_regulation: "foo"
+# alpha: 123.0
+# q: 123.0
+estimator: "mckp"
+estimator_multiple_cpu: false
+# constant_learning_rate: true
+debug: false
+cuda: false
+
+# Nextflow input-output arguments
+publish_dir: # please fill in - example: "output/"
+# param_list: "my_params.yaml"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_schema.json b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_schema.json
new file mode 100644
index 0000000..03e3ac9
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/nextflow_schema.json
@@ -0,0 +1,551 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "cellbender_remove_background",
+"description": "Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n",
+"type": "object",
+"definitions": {
+
+    
+    "Dataset input": {
+        "title": "Dataset input",
+        "type": "object",
+        "description": "Dataset input using nf-tower \"dataset\" or \"data explorer\". Allows for the input of multiple         parameter sets to initialise a Nextflow channel.",
+        "properties": {
+            "param_list": {
+                "description": "Dataset input can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml                 blob. The names of the input fields (e.g. csv columns, json keys) need to be an exact match with the workflow input parameters.",
+                "default": "",
+                "format": "file-path",
+                "mimetype": "text/csv",
+                "pattern": "^\\S+\\.csv$"
+            }
+        }
+    },
+
+    
+    
+    "inputs" : {
+    "title": "Inputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, example: `input.h5mu`. Input h5mu file",
+                "help_text": "Type: `file`, required, example: `input.h5mu`. Input h5mu file. Data file on which to run tool. Data must be un-filtered: it should include empty droplets."
+            
+            }
+    
+
+        ,
+                "modality": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `rna`. List of modalities to process",
+                "help_text": "Type: `string`, default: `rna`. List of modalities to process."
+            ,
+                "default":"rna"
+            }
+    
+
+}
+},
+    
+    
+    "outputs" : {
+    "title": "Outputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Full count matrix as an h5mu file, with background RNA removed",
+                "help_text": "Type: `file`, required, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Full count matrix as an h5mu file, with background RNA removed. This file contains all the original droplet barcodes."
+            ,
+                "default":"$id.$key.output.h5mu"
+            }
+    
+
+        ,
+                "output_compression": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. ",
+                "help_text": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. ",
+                "enum": ["gzip", "lzf"]
+            
+            
+            }
+    
+
+        ,
+                "layer_output": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_corrected`. Output layer",
+                "help_text": "Type: `string`, default: `cellbender_corrected`. Output layer"
+            ,
+                "default":"cellbender_corrected"
+            }
+    
+
+        ,
+                "obs_background_fraction": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_background_fraction`. ",
+                "help_text": "Type: `string`, default: `cellbender_background_fraction`. "
+            ,
+                "default":"cellbender_background_fraction"
+            }
+    
+
+        ,
+                "obs_cell_probability": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_cell_probability`. ",
+                "help_text": "Type: `string`, default: `cellbender_cell_probability`. "
+            ,
+                "default":"cellbender_cell_probability"
+            }
+    
+
+        ,
+                "obs_cell_size": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_cell_size`. ",
+                "help_text": "Type: `string`, default: `cellbender_cell_size`. "
+            ,
+                "default":"cellbender_cell_size"
+            }
+    
+
+        ,
+                "obs_droplet_efficiency": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_droplet_efficiency`. ",
+                "help_text": "Type: `string`, default: `cellbender_droplet_efficiency`. "
+            ,
+                "default":"cellbender_droplet_efficiency"
+            }
+    
+
+        ,
+                "obs_latent_scale": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_latent_scale`. ",
+                "help_text": "Type: `string`, default: `cellbender_latent_scale`. "
+            ,
+                "default":"cellbender_latent_scale"
+            }
+    
+
+        ,
+                "var_ambient_expression": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_ambient_expression`. ",
+                "help_text": "Type: `string`, default: `cellbender_ambient_expression`. "
+            ,
+                "default":"cellbender_ambient_expression"
+            }
+    
+
+        ,
+                "obsm_gene_expression_encoding": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `cellbender_gene_expression_encoding`. ",
+                "help_text": "Type: `string`, default: `cellbender_gene_expression_encoding`. "
+            ,
+                "default":"cellbender_gene_expression_encoding"
+            }
+    
+
+}
+},
+    
+    
+    "arguments" : {
+    "title": "Arguments",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "expected_cells_from_qc": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean`, default: `false`. Will use the Cell Ranger QC to determine the estimated number of cells",
+                "help_text": "Type: `boolean`, default: `false`. Will use the Cell Ranger QC to determine the estimated number of cells"
+            ,
+                "default":false
+            }
+    
+
+        ,
+                "expected_cells": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, example: `1000`. Number of cells expected in the dataset (a rough estimate within a factor of 2 is sufficient)",
+                "help_text": "Type: `integer`, example: `1000`. Number of cells expected in the dataset (a rough estimate within a factor of 2 is sufficient)."
+            
+            }
+    
+
+        ,
+                "total_droplets_included": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, example: `25000`. The number of droplets from the rank-ordered UMI plot\nthat will have their cell probabilities inferred as an\noutput",
+                "help_text": "Type: `integer`, example: `25000`. The number of droplets from the rank-ordered UMI plot\nthat will have their cell probabilities inferred as an\noutput. Include the droplets which might contain cells.\nDroplets beyond TOTAL_DROPLETS_INCLUDED should be\n\u0027surely empty\u0027 droplets.\n"
+            
+            }
+    
+
+        ,
+                "force_cell_umi_prior": {
+                "type":
+                "integer",
+                "description": "Type: `integer`. Ignore CellBender\u0027s heuristic prior estimation, and use this prior for UMI counts in cells",
+                "help_text": "Type: `integer`. Ignore CellBender\u0027s heuristic prior estimation, and use this prior for UMI counts in cells."
+            
+            }
+    
+
+        ,
+                "force_empty_umi_prior": {
+                "type":
+                "integer",
+                "description": "Type: `integer`. Ignore CellBender\u0027s heuristic prior estimation, and use this prior for UMI counts in empty droplets",
+                "help_text": "Type: `integer`. Ignore CellBender\u0027s heuristic prior estimation, and use this prior for UMI counts in empty droplets."
+            
+            }
+    
+
+        ,
+                "model": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `full`, choices: ``naive`, `simple`, `ambient`, `swapping`, `full``. Which model is being used for count data",
+                "help_text": "Type: `string`, default: `full`, choices: ``naive`, `simple`, `ambient`, `swapping`, `full``. Which model is being used for count data.\n\n* \u0027naive\u0027 subtracts the estimated ambient profile.\n* \u0027simple\u0027 does not model either ambient RNA or random barcode swapping (for debugging purposes -- not recommended).\n* \u0027ambient\u0027 assumes background RNA is incorporated into droplets.\n* \u0027swapping\u0027 assumes background RNA comes from random barcode swapping (via PCR chimeras).\n* \u0027full\u0027 uses a combined ambient and swapping model.\n",
+                "enum": ["naive", "simple", "ambient", "swapping", "full"]
+            
+            ,
+                "default":"full"
+            }
+    
+
+        ,
+                "epochs": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, default: `150`. Number of epochs to train",
+                "help_text": "Type: `integer`, default: `150`. Number of epochs to train."
+            ,
+                "default":150
+            }
+    
+
+        ,
+                "low_count_threshold": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, default: `5`. Droplets with UMI counts below this number are completely \nexcluded from the analysis",
+                "help_text": "Type: `integer`, default: `5`. Droplets with UMI counts below this number are completely \nexcluded from the analysis. This can help identify the correct \nprior for empty droplet counts in the rare case where empty \ncounts are extremely high (over 200).\n"
+            ,
+                "default":5
+            }
+    
+
+        ,
+                "z_dim": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, default: `64`. Dimension of latent variable z",
+                "help_text": "Type: `integer`, default: `64`. Dimension of latent variable z.\n"
+            ,
+                "default":64
+            }
+    
+
+        ,
+                "z_layers": {
+                "type":
+                "string",
+                "description": "Type: List of `integer`, default: `512`, multiple_sep: `\";\"`. Dimension of hidden layers in the encoder for z",
+                "help_text": "Type: List of `integer`, default: `512`, multiple_sep: `\";\"`. Dimension of hidden layers in the encoder for z.\n"
+            ,
+                "default":"512"
+            }
+    
+
+        ,
+                "training_fraction": {
+                "type":
+                "number",
+                "description": "Type: `double`, default: `0.9`. Training detail: the fraction of the data used for training",
+                "help_text": "Type: `double`, default: `0.9`. Training detail: the fraction of the data used for training.\nThe rest is never seen by the inference algorithm. Speeds up learning.\n"
+            ,
+                "default":0.9
+            }
+    
+
+        ,
+                "empty_drop_training_fraction": {
+                "type":
+                "number",
+                "description": "Type: `double`, default: `0.2`. Training detail: the fraction of the training data each epoch that \nis drawn (randomly sampled) from surely empty droplets",
+                "help_text": "Type: `double`, default: `0.2`. Training detail: the fraction of the training data each epoch that \nis drawn (randomly sampled) from surely empty droplets.\n"
+            ,
+                "default":0.2
+            }
+    
+
+        ,
+                "ignore_features": {
+                "type":
+                "string",
+                "description": "Type: List of `integer`, multiple_sep: `\";\"`. Integer indices of features to ignore entirely",
+                "help_text": "Type: List of `integer`, multiple_sep: `\";\"`. Integer indices of features to ignore entirely. In the output\ncount matrix, the counts for these features will be unchanged.\n"
+            
+            }
+    
+
+        ,
+                "fpr": {
+                "type":
+                "string",
+                "description": "Type: List of `double`, default: `0.01`, multiple_sep: `\";\"`. Target \u0027delta\u0027 false positive rate in [0, 1)",
+                "help_text": "Type: List of `double`, default: `0.01`, multiple_sep: `\";\"`. Target \u0027delta\u0027 false positive rate in [0, 1). Use 0 for a cohort\nof samples which will be jointly analyzed for differential expression.\nA false positive is a true signal count that is erroneously removed.\nMore background removal is accompanied by more signal removal at\nhigh values of FPR. You can specify multiple values, which will\ncreate multiple output files.\n"
+            ,
+                "default":"0.01"
+            }
+    
+
+        ,
+                "exclude_feature_types": {
+                "type":
+                "string",
+                "description": "Type: List of `string`, multiple_sep: `\";\"`. Feature types to ignore during the analysis",
+                "help_text": "Type: List of `string`, multiple_sep: `\";\"`. Feature types to ignore during the analysis. These features will\nbe left unchanged in the output file.\n"
+            
+            }
+    
+
+        ,
+                "projected_ambient_count_threshold": {
+                "type":
+                "number",
+                "description": "Type: `double`, default: `0.1`. Controls how many features are included in the analysis, which\ncan lead to a large speedup",
+                "help_text": "Type: `double`, default: `0.1`. Controls how many features are included in the analysis, which\ncan lead to a large speedup. If a feature is expected to have less\nthan PROJECTED_AMBIENT_COUNT_THRESHOLD counts total in all cells\n(summed), then that gene is excluded, and it will be unchanged\nin the output count matrix. For example, \nPROJECTED_AMBIENT_COUNT_THRESHOLD = 0 will include all features\nwhich have even a single count in any empty droplet.\n"
+            ,
+                "default":0.1
+            }
+    
+
+        ,
+                "learning_rate": {
+                "type":
+                "number",
+                "description": "Type: `double`, default: `1.0E-4`. Training detail: lower learning rate for inference",
+                "help_text": "Type: `double`, default: `1.0E-4`. Training detail: lower learning rate for inference.\nA OneCycle learning rate schedule is used, where the\nupper learning rate is ten times this value. (For this\nvalue, probably do not exceed 1e-3).\n"
+            ,
+                "default":0.0001
+            }
+    
+
+        ,
+                "final_elbo_fail_fraction": {
+                "type":
+                "number",
+                "description": "Type: `double`. Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO - initial_test_ELBO) \u003e FINAL_ELBO_FAIL_FRACTION",
+                "help_text": "Type: `double`. Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO - initial_test_ELBO) \u003e FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries \u003e 1.\nBy default, will not fail training based on final_training_ELBO.\n"
+            
+            }
+    
+
+        ,
+                "epoch_elbo_fail_fraction": {
+                "type":
+                "number",
+                "description": "Type: `double`. Training is considered to have failed if \n(previous_epoch_test_ELBO - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO) \u003e EPOCH_ELBO_FAIL_FRACTION",
+                "help_text": "Type: `double`. Training is considered to have failed if \n(previous_epoch_test_ELBO - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO) \u003e EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries \u003e 1.\nBy default, will not fail training based on epoch_training_ELBO.\n"
+            
+            }
+    
+
+        ,
+                "num_training_tries": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, default: `1`. Number of times to attempt to train the model",
+                "help_text": "Type: `integer`, default: `1`. Number of times to attempt to train the model. At each subsequent attempt,\nthe learning rate is multiplied by LEARNING_RATE_RETRY_MULT.\n"
+            ,
+                "default":1
+            }
+    
+
+        ,
+                "learning_rate_retry_mult": {
+                "type":
+                "number",
+                "description": "Type: `double`, default: `0.2`. Learning rate is multiplied by this amount each time a new training\nattempt is made",
+                "help_text": "Type: `double`, default: `0.2`. Learning rate is multiplied by this amount each time a new training\nattempt is made. (This parameter is only used if training fails based\non EPOCH_ELBO_FAIL_FRACTION or FINAL_ELBO_FAIL_FRACTION and\nNUM_TRAINING_TRIES is \u003e 1.) \n"
+            ,
+                "default":0.2
+            }
+    
+
+        ,
+                "posterior_batch_size": {
+                "type":
+                "integer",
+                "description": "Type: `integer`, default: `128`. Training detail: size of batches when creating the posterior",
+                "help_text": "Type: `integer`, default: `128`. Training detail: size of batches when creating the posterior.\nReduce this to avoid running out of GPU memory creating the posterior\n(will be slower).\n"
+            ,
+                "default":128
+            }
+    
+
+        ,
+                "posterior_regulation": {
+                "type":
+                "string",
+                "description": "Type: `string`, choices: ``PRq`, `PRmu`, `PRmu_gene``. Posterior regularization method",
+                "help_text": "Type: `string`, choices: ``PRq`, `PRmu`, `PRmu_gene``. Posterior regularization method. (For experts: not required for normal usage,\nsee documentation). \n\n* PRq is approximate quantile-targeting.\n* PRmu is approximate mean-targeting aggregated over genes (behavior of v0.2.0).\n* PRmu_gene is approximate mean-targeting per gene.\n",
+                "enum": ["PRq", "PRmu", "PRmu_gene"]
+            
+            
+            }
+    
+
+        ,
+                "alpha": {
+                "type":
+                "number",
+                "description": "Type: `double`. Tunable parameter alpha for the PRq posterior regularization method\n(not normally used: see documentation)",
+                "help_text": "Type: `double`. Tunable parameter alpha for the PRq posterior regularization method\n(not normally used: see documentation).\n"
+            
+            }
+    
+
+        ,
+                "q": {
+                "type":
+                "number",
+                "description": "Type: `double`. Tunable parameter q for the CDF threshold estimation method (not\nnormally used: see documentation)",
+                "help_text": "Type: `double`. Tunable parameter q for the CDF threshold estimation method (not\nnormally used: see documentation).\n"
+            
+            }
+    
+
+        ,
+                "estimator": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `mckp`, choices: ``map`, `mean`, `cdf`, `sample`, `mckp``. Output denoised count estimation method",
+                "help_text": "Type: `string`, default: `mckp`, choices: ``map`, `mean`, `cdf`, `sample`, `mckp``. Output denoised count estimation method. (For experts: not required\nfor normal usage, see documentation).\n",
+                "enum": ["map", "mean", "cdf", "sample", "mckp"]
+            
+            ,
+                "default":"mckp"
+            }
+    
+
+        ,
+                "estimator_multiple_cpu": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean_true`, default: `false`. Including the flag --estimator-multiple-cpu will use more than one\nCPU to compute the MCKP output count estimator in parallel (does nothing\nfor other estimators)",
+                "help_text": "Type: `boolean_true`, default: `false`. Including the flag --estimator-multiple-cpu will use more than one\nCPU to compute the MCKP output count estimator in parallel (does nothing\nfor other estimators).\n"
+            ,
+                "default":false
+            }
+    
+
+        ,
+                "constant_learning_rate": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean`. Including the flag --constant-learning-rate will use the ClippedAdam\noptimizer instead of the OneCycleLR learning rate schedule, which is\nthe default",
+                "help_text": "Type: `boolean`. Including the flag --constant-learning-rate will use the ClippedAdam\noptimizer instead of the OneCycleLR learning rate schedule, which is\nthe default. Learning is faster with the OneCycleLR schedule.\nHowever, training can easily be continued from a checkpoint for more\nepochs than the initial command specified when using ClippedAdam. On\nthe other hand, if using the OneCycleLR schedule with 150 epochs\nspecified, it is not possible to pick up from that final checkpoint\nand continue training until 250 epochs.\n"
+            
+            }
+    
+
+        ,
+                "debug": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean_true`, default: `false`. Including the flag --debug will log extra messages useful for debugging",
+                "help_text": "Type: `boolean_true`, default: `false`. Including the flag --debug will log extra messages useful for debugging.\n"
+            ,
+                "default":false
+            }
+    
+
+        ,
+                "cuda": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean_true`, default: `false`. Including the flag --cuda will run the inference on a\nGPU",
+                "help_text": "Type: `boolean_true`, default: `false`. Including the flag --cuda will run the inference on a\nGPU.\n"
+            ,
+                "default":false
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/inputs"
+    },
+
+    {
+    "$ref": "#/definitions/outputs"
+    },
+
+    {
+    "$ref": "#/definitions/arguments"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/setup_logger.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/.config.vsh.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/.config.vsh.yaml
new file mode 100644
index 0000000..0f60f47
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/.config.vsh.yaml
@@ -0,0 +1,262 @@
+name: "add_id"
+namespace: "metadata"
+version: "2.1.2"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_id"
+    description: "The input id."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output"
+    description: "Name of the .obs column where to store the id."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "The compression format to be used on the output h5mu object."
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--make_observation_keys_unique"
+    description: "Join the id to the .obs index (.obs_names)."
+    info: null
+    direction: "input"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add id of .obs. Also allows to make .obs_names (the .obs index) unique\
+  \ \nby prefixing the values with an unique id per .h5mu file.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_tag: "2.1.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+build_info:
+  config: "src/metadata/add_id/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker"
+  output: "target/nextflow/metadata/add_id"
+  executable: "target/nextflow/metadata/add_id/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "a0c9522486585774f76416150f8a3291409b5363"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  git_tag: "2.1.1-2-ga0c95224865"
+package_config:
+  name: "openpipeline"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".version := \"2.1.2\""
+  - ".engines[.type == 'docker'].target_tag := '2.1.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "openpipelines-bio"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/main.nf b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/main.nf
new file mode 100644
index 0000000..e33df32
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/main.nf
@@ -0,0 +1,3983 @@
+// add_id 2.1.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "add_id",
+  "namespace" : "metadata",
+  "version" : "2.1.2",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_id",
+          "description" : "The input id.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output",
+          "description" : "Name of the .obs column where to store the id.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "The compression format to be used on the output h5mu object.",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--make_observation_keys_unique",
+          "description" : "Join the id to the .obs index (.obs_names).",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_tag" : "2.1.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    }
+  ],
+  "build_info" : {
+    "config" : "/home/runner/work/openpipeline/openpipeline/src/metadata/add_id/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker",
+    "output" : "/home/runner/work/openpipeline/openpipeline/target/nextflow/metadata/add_id",
+    "viash_version" : "0.9.4",
+    "git_commit" : "a0c9522486585774f76416150f8a3291409b5363",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline",
+    "git_tag" : "2.1.1-2-ga0c95224865"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/home/runner/work/openpipeline/openpipeline/src",
+    "target" : "/home/runner/work/openpipeline/openpipeline/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".version := \\"2.1.2\\"",
+      ".engines[.type == 'docker'].target_tag := '2.1.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "openpipelines-bio",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+from mudata import read_h5mu, MuData
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_id': $( if [ ! -z ${VIASH_PAR_INPUT_ID+x} ]; then echo "r'${VIASH_PAR_INPUT_ID//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'make_observation_keys_unique': $( if [ ! -z ${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE+x} ]; then echo "r'${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def make_observation_keys_unique(sample_id: str, sample: MuData) -> None:
+    """
+    Make the observation keys unique across all samples. At input,
+    the observation keys are unique within a sample. By adding the sample name
+    (unique for a sample) to each observation key, the observation key is made
+    unique across all samples as well.
+    """
+    logger.info(
+        "Making observation keys unique across all "
+        "samples by appending prefix '%s' to the observation names.",
+        sample_id,
+    )
+    sample.obs.index = f"{sample_id}_" + sample.obs.index
+    make_observation_keys_unique_per_mod(sample_id, sample)
+    logger.info("Done making observation keys unique.")
+
+
+def make_observation_keys_unique_per_mod(sample_id: str, sample: MuData) -> None:
+    """
+    Updating MuData.obs_names is not allowed (it is read-only).
+    So the observation keys for each modality has to be updated manually.
+    """
+    for mod_name, mod in sample.mod.items():
+        logger.info("Processing modality '%s'", mod_name)
+        mod.obs_names = f"{sample_id}_" + mod.obs_names
+
+
+def main():
+    logger.info("Reading input file '%s'.", par["input"])
+    input_data = read_h5mu(par["input"])
+    logger.info(
+        "Adding column '%s' to global .obs dataframe, populated with ID '%s'",
+        par["obs_output"],
+        par["input_id"],
+    )
+    input_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding column to global .obs")
+    for mod_name, mod_data in input_data.mod.items():
+        logger.info(
+            "Adding column '%s' to .obs dataframe for modality '%s', "
+            "populated with ID '%s'",
+            par["obs_output"],
+            mod_name,
+            par["input_id"],
+        )
+        mod_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding per-modality columns.")
+    if par["make_observation_keys_unique"]:
+        make_observation_keys_unique(par["input_id"], input_data)
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        par["output"],
+        par["output_compression"],
+    )
+    input_data.write_h5mu(par["output"], compression=par["output_compression"])
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "ghcr.io",
+    "image" : "openpipelines-bio/openpipeline/metadata/add_id",
+    "tag" : "2.1.0"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow.config
new file mode 100644
index 0000000..d706fdc
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/add_id'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = '2.1.2'
+  description = 'Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_params.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_params.yaml
new file mode 100644
index 0000000..9be66e5
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_params.yaml
@@ -0,0 +1,11 @@
+# Arguments
+input: # please fill in - example: "sample_path"
+input_id: # please fill in - example: "foo"
+obs_output: "sample_id"
+# output: "$id.$key.output.h5mu"
+# output_compression: "gzip"
+make_observation_keys_unique: false
+
+# Nextflow input-output arguments
+publish_dir: # please fill in - example: "output/"
+# param_list: "my_params.yaml"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_schema.json b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_schema.json
new file mode 100644
index 0000000..a12362a
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/nextflow_schema.json
@@ -0,0 +1,132 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "add_id",
+"description": "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+"type": "object",
+"definitions": {
+
+    
+    "Dataset input": {
+        "title": "Dataset input",
+        "type": "object",
+        "description": "Dataset input using nf-tower \"dataset\" or \"data explorer\". Allows for the input of multiple         parameter sets to initialise a Nextflow channel.",
+        "properties": {
+            "param_list": {
+                "description": "Dataset input can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml                 blob. The names of the input fields (e.g. csv columns, json keys) need to be an exact match with the workflow input parameters.",
+                "default": "",
+                "format": "file-path",
+                "mimetype": "text/csv",
+                "pattern": "^\\S+\\.csv$"
+            }
+        }
+    },
+
+    
+    
+    "arguments" : {
+    "title": "Arguments",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, example: `sample_path`. Path to the input ",
+                "help_text": "Type: `file`, required, example: `sample_path`. Path to the input .h5mu."
+            
+            }
+    
+
+        ,
+                "input_id": {
+                "type":
+                "string",
+                "description": "Type: `string`, required. The input id",
+                "help_text": "Type: `string`, required. The input id."
+            
+            }
+    
+
+        ,
+                "obs_output": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `sample_id`. Name of the ",
+                "help_text": "Type: `string`, default: `sample_id`. Name of the .obs column where to store the id."
+            ,
+                "default":"sample_id"
+            }
+    
+
+        ,
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. ",
+                "help_text": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. "
+            ,
+                "default":"$id.$key.output.h5mu"
+            }
+    
+
+        ,
+                "output_compression": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object",
+                "help_text": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object.",
+                "enum": ["gzip", "lzf"]
+            
+            
+            }
+    
+
+        ,
+                "make_observation_keys_unique": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean_true`, default: `false`. Join the id to the ",
+                "help_text": "Type: `boolean_true`, default: `false`. Join the id to the .obs index (.obs_names)."
+            ,
+                "default":false
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/arguments"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/setup_logger.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/.config.vsh.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
new file mode 100644
index 0000000..f0f6adc
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
@@ -0,0 +1,318 @@
+name: "grep_annotation_column"
+namespace: "metadata"
+version: "2.1.2"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  description: "Arguments related to the input dataset."
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_column"
+    description: "Column to query. If not specified, use .var_names or .obs_names,\
+      \ depending on the value of --matrix"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input data to use when calculating fraction of observations that\
+      \ match with the query. \nOnly used when --output_fraction_column is provided.\
+      \ If not specified, .X is used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to get the annotation matrix from.\n"
+    info: null
+    example:
+    - "rna"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--matrix"
+    description: "Matrix to fetch the column from that will be searched."
+    info: null
+    example:
+    - "var"
+    required: false
+    choices:
+    - "var"
+    - "obs"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "Arguments related to how the output will be written."
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "The compression format to be used on the output h5mu object."
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_match_column"
+    description: "Name of the column to write the result to."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_fraction_column"
+    description: "For the opposite axis, name of the column to write the fraction\
+      \ of \nobservations that matches to the pattern.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Query options"
+  description: "Options related to the query"
+  arguments:
+  - type: "string"
+    name: "--regex_pattern"
+    description: "Regex to use to match with the input column."
+    info: null
+    example:
+    - "^[mM][tT]-"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Perform a regex lookup on a column from the annotation matrices .obs\
+  \ or .var.\nThe annotation matrix can originate from either a modality, or all modalities\
+  \ (global .var or .obs).\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_tag: "2.1.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+build_info:
+  config: "src/metadata/grep_annotation_column/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker"
+  output: "target/nextflow/metadata/grep_annotation_column"
+  executable: "target/nextflow/metadata/grep_annotation_column/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "a0c9522486585774f76416150f8a3291409b5363"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  git_tag: "2.1.1-2-ga0c95224865"
+package_config:
+  name: "openpipeline"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".version := \"2.1.2\""
+  - ".engines[.type == 'docker'].target_tag := '2.1.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "openpipelines-bio"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/compress_h5mu.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/main.nf b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/main.nf
new file mode 100644
index 0000000..f9d4bdc
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/main.nf
@@ -0,0 +1,4122 @@
+// grep_annotation_column 2.1.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "grep_annotation_column",
+  "namespace" : "metadata",
+  "version" : "2.1.2",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "description" : "Arguments related to the input dataset.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_column",
+          "description" : "Column to query. If not specified, use .var_names or .obs_names, depending on the value of --matrix",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input data to use when calculating fraction of observations that match with the query. \nOnly used when --output_fraction_column is provided. If not specified, .X is used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to get the annotation matrix from.\n",
+          "example" : [
+            "rna"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--matrix",
+          "description" : "Matrix to fetch the column from that will be searched.",
+          "example" : [
+            "var"
+          ],
+          "required" : false,
+          "choices" : [
+            "var",
+            "obs"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "Arguments related to how the output will be written.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "The compression format to be used on the output h5mu object.",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_match_column",
+          "description" : "Name of the column to write the result to.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_fraction_column",
+          "description" : "For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Query options",
+      "description" : "Options related to the query",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--regex_pattern",
+          "description" : "Regex to use to match with the input column.",
+          "example" : [
+            "^[mM][tT]-"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_tag" : "2.1.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    }
+  ],
+  "build_info" : {
+    "config" : "/home/runner/work/openpipeline/openpipeline/src/metadata/grep_annotation_column/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker",
+    "output" : "/home/runner/work/openpipeline/openpipeline/target/nextflow/metadata/grep_annotation_column",
+    "viash_version" : "0.9.4",
+    "git_commit" : "a0c9522486585774f76416150f8a3291409b5363",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline",
+    "git_tag" : "2.1.1-2-ga0c95224865"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/home/runner/work/openpipeline/openpipeline/src",
+    "target" : "/home/runner/work/openpipeline/openpipeline/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".version := \\"2.1.2\\"",
+      ".engines[.type == 'docker'].target_tag := '2.1.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "openpipelines-bio",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata as mu
+from pathlib import Path
+from operator import attrgetter
+from pandas import Series
+import scipy as sc
+import re
+import numpy as np
+
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_column': $( if [ ! -z ${VIASH_PAR_INPUT_COLUMN+x} ]; then echo "r'${VIASH_PAR_INPUT_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'matrix': $( if [ ! -z ${VIASH_PAR_MATRIX+x} ]; then echo "r'${VIASH_PAR_MATRIX//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_match_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_MATCH_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MATCH_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_fraction_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_FRACTION_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_FRACTION_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'regex_pattern': $( if [ ! -z ${VIASH_PAR_REGEX_PATTERN+x} ]; then echo "r'${VIASH_PAR_REGEX_PATTERN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def describe_array(arr, msg):
+    # Note: sc.stats returns a DescribeResult NamedTuple. For NamedTuples,
+    # the _asdict method is public facing even though it starts with an underscore.
+    description = sc.stats.describe(arr)._asdict()
+    logger.info(
+        "%s:\\\\nshape: %s\\\\nmean: %s\\\\nnobs: %s\\\\n"
+        "variance: %s\\\\nmin: %s\\\\nmax: %s\\\\ncontains na: %s\\\\ndtype: %s\\\\ncontains 0: %s",
+        msg,
+        arr.shape,
+        description["mean"],
+        description["nobs"],
+        description["variance"],
+        description["minmax"][0],
+        description["minmax"][1],
+        np.isnan(arr).any(),
+        arr.dtype,
+        (arr == 0).any(),
+    )
+
+
+def main(par):
+    input_file, output_file, mod_name = (
+        Path(par["input"]),
+        Path(par["output"]),
+        par["modality"],
+    )
+    logger.info(f"Compiling regular expression '{par['regex_pattern']}'.")
+    try:
+        compiled_regex = re.compile(par["regex_pattern"])
+    except (TypeError, re.error) as e:
+        raise ValueError(
+            f"{par['regex_pattern']} is not a valid regular expression pattern."
+        ) from e
+    else:
+        if compiled_regex.groups:
+            raise NotImplementedError(
+                "Using match groups is not supported by this component."
+            )
+    logger.info("Reading input file %s, modality %s.", input_file, mod_name)
+
+    modality_data = mu.read_h5ad(input_file, mod=mod_name)
+    logger.info("Reading input file done.")
+    logger.info("Using annotation dataframe '%s'.", par["matrix"])
+    annotation_matrix = getattr(modality_data, par["matrix"])
+    default_column = {"var": attrgetter("var_names"), "obs": attrgetter("obs_names")}
+    if par["input_column"]:
+        logger.info("Input column '%s' was specified.", par["input_column"])
+        try:
+            annotation_column = annotation_matrix[par["input_column"]]
+        except KeyError as e:
+            raise ValueError(
+                f"Column {par['input_column']} could not be found for modality "
+                f"{par['modality']}. Available columns:"
+                f" {','.join(annotation_matrix.columns.to_list())}"
+            ) from e
+    else:
+        logger.info(f"No input column specified, using '.{par['matrix']}_names'")
+        annotation_column = default_column[par["matrix"]](modality_data).to_series()
+    logger.info("Applying regex search.")
+    grep_result = annotation_column.str.contains(par["regex_pattern"], regex=True)
+    logger.info("Search results: %s", grep_result.value_counts())
+    # A Series object cannot be used as an indexer for a scipy sparse array
+    # when the data type is a pandas boolean extension array because
+    # extension arrays do not define .nonzero()
+    # See https://github.com/pandas-dev/pandas/issues/46025
+    grep_result = grep_result.to_numpy(dtype="bool", na_value=False)
+
+    other_axis_attribute = {"var": "obs", "obs": "var"}
+    if par["output_fraction_column"]:
+        logger.info(
+            "Enabled writing the fraction of values that matches to the pattern."
+        )
+        input_layer = (
+            modality_data.X
+            if not par["input_layer"]
+            else modality_data.layers[par["input_layer"]]
+        )
+        totals = np.ravel(input_layer.sum(axis=1))
+        describe_array(totals, "Summary of total counts for layer")
+        counts_for_matches = np.ravel(input_layer[:, grep_result].sum(axis=1))
+        describe_array(counts_for_matches, "Summary of counts matching grep")
+        with np.errstate(all="raise"):
+            pct_matching = np.divide(
+                counts_for_matches,
+                totals,
+                out=np.zeros_like(totals, dtype=np.float64),
+                where=(~np.isclose(totals, np.zeros_like(totals))),
+            )
+        logger.info("Testing wether or not fractions data contains NA.")
+        assert ~np.isnan(pct_matching).any(), "Fractions should not contain NA."
+        logger.info("Fraction statistics: \\\\n%s", Series(pct_matching).describe())
+        pct_matching = np.where(np.isclose(pct_matching, 0, atol=1e-6), 0, pct_matching)
+        pct_matching = np.where(np.isclose(pct_matching, 1, atol=1e-6), 1, pct_matching)
+        assert (
+            np.logical_and(pct_matching >= 0, pct_matching <= 1)
+        ).all(), "Fractions are not within bounds, please report this as a bug"
+        output_matrix = other_axis_attribute[par["matrix"]]
+        logger.info(
+            "Writing fractions to matrix '%s', column '%s'",
+            output_matrix,
+            par["output_fraction_column"],
+        )
+        getattr(modality_data, output_matrix)[par["output_fraction_column"]] = (
+            pct_matching
+        )
+    logger.info(
+        "Adding values that matched the pattern to '%s', column '%s'",
+        par["matrix"],
+        par["output_match_column"],
+    )
+    getattr(modality_data, par["matrix"])[par["output_match_column"]] = grep_result
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        output_file,
+        par["output_compression"],
+    )
+    write_h5ad_to_h5mu_with_compression(
+        output_file, par["input"], mod_name, modality_data, par["output_compression"]
+    )
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "ghcr.io",
+    "image" : "openpipelines-bio/openpipeline/metadata/grep_annotation_column",
+    "tag" : "2.1.0"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow.config
new file mode 100644
index 0000000..52ff2cb
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/grep_annotation_column'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = '2.1.2'
+  description = 'Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_params.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_params.yaml
new file mode 100644
index 0000000..ad24b3d
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_params.yaml
@@ -0,0 +1,21 @@
+# Inputs
+input: # please fill in - example: "sample_path"
+# input_column: "foo"
+# input_layer: "foo"
+modality: # please fill in - example: "rna"
+# matrix: "var"
+
+# Outputs
+# output: "$id.$key.output.h5mu"
+# output_compression: "gzip"
+output_match_column: # please fill in - example: "foo"
+# output_fraction_column: "foo"
+
+# Query options
+regex_pattern: # please fill in - example: "^[mM][tT]-"
+
+# Nextflow input-output arguments
+publish_dir: # please fill in - example: "output/"
+# param_list: "my_params.yaml"
+
+# Arguments
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_schema.json b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_schema.json
new file mode 100644
index 0000000..799aa8b
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/nextflow_schema.json
@@ -0,0 +1,200 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "grep_annotation_column",
+"description": "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+"type": "object",
+"definitions": {
+
+    
+    "Dataset input": {
+        "title": "Dataset input",
+        "type": "object",
+        "description": "Dataset input using nf-tower \"dataset\" or \"data explorer\". Allows for the input of multiple         parameter sets to initialise a Nextflow channel.",
+        "properties": {
+            "param_list": {
+                "description": "Dataset input can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml                 blob. The names of the input fields (e.g. csv columns, json keys) need to be an exact match with the workflow input parameters.",
+                "default": "",
+                "format": "file-path",
+                "mimetype": "text/csv",
+                "pattern": "^\\S+\\.csv$"
+            }
+        }
+    },
+
+    
+    
+    "inputs" : {
+    "title": "Inputs",
+    "type": "object",
+    "description": "Arguments related to the input dataset.",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, example: `sample_path`. Path to the input ",
+                "help_text": "Type: `file`, required, example: `sample_path`. Path to the input .h5mu."
+            
+            }
+    
+
+        ,
+                "input_column": {
+                "type":
+                "string",
+                "description": "Type: `string`. Column to query",
+                "help_text": "Type: `string`. Column to query. If not specified, use .var_names or .obs_names, depending on the value of --matrix"
+            
+            }
+    
+
+        ,
+                "input_layer": {
+                "type":
+                "string",
+                "description": "Type: `string`. Input data to use when calculating fraction of observations that match with the query",
+                "help_text": "Type: `string`. Input data to use when calculating fraction of observations that match with the query. \nOnly used when --output_fraction_column is provided. If not specified, .X is used.\n"
+            
+            }
+    
+
+        ,
+                "modality": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `rna`. Which modality to get the annotation matrix from",
+                "help_text": "Type: `string`, required, example: `rna`. Which modality to get the annotation matrix from.\n"
+            
+            }
+    
+
+        ,
+                "matrix": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `var`, choices: ``var`, `obs``. Matrix to fetch the column from that will be searched",
+                "help_text": "Type: `string`, example: `var`, choices: ``var`, `obs``. Matrix to fetch the column from that will be searched.",
+                "enum": ["var", "obs"]
+            
+            
+            }
+    
+
+}
+},
+    
+    
+    "outputs" : {
+    "title": "Outputs",
+    "type": "object",
+    "description": "Arguments related to how the output will be written.",
+    "properties": {
+    
+        
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. ",
+                "help_text": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. "
+            ,
+                "default":"$id.$key.output.h5mu"
+            }
+    
+
+        ,
+                "output_compression": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object",
+                "help_text": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object.",
+                "enum": ["gzip", "lzf"]
+            
+            
+            }
+    
+
+        ,
+                "output_match_column": {
+                "type":
+                "string",
+                "description": "Type: `string`, required. Name of the column to write the result to",
+                "help_text": "Type: `string`, required. Name of the column to write the result to."
+            
+            }
+    
+
+        ,
+                "output_fraction_column": {
+                "type":
+                "string",
+                "description": "Type: `string`. For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern",
+                "help_text": "Type: `string`. For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n"
+            
+            }
+    
+
+}
+},
+    
+    
+    "query options" : {
+    "title": "Query options",
+    "type": "object",
+    "description": "Options related to the query",
+    "properties": {
+    
+        
+                "regex_pattern": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `^[mM][tT]-`. Regex to use to match with the input column",
+                "help_text": "Type: `string`, required, example: `^[mM][tT]-`. Regex to use to match with the input column."
+            
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/inputs"
+    },
+
+    {
+    "$ref": "#/definitions/outputs"
+    },
+
+    {
+    "$ref": "#/definitions/query options"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/setup_logger.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/grep_annotation_column/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
new file mode 100644
index 0000000..5b2d5f8
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
@@ -0,0 +1,376 @@
+name: "calculate_qc_metrics"
+namespace: "qc"
+version: "2.1.2"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .obs"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable,mitochondrial"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--var_qc_metrics_fill_na_value"
+    description: "Fill any 'NA' values found in the columns specified with --var_qc_metrics\
+      \ to 'True' or 'False'.\nas False.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20;50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .var"
+  arguments:
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "The compression format to be used on the output h5mu object."
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are\
+  \ comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have\
+  \ slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n\
+  \  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n\
+  \  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs\
+  \ metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics}\
+  \ -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n\
+  \    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n\
+  \    - total_counts -> total_{expr_type}\n    \n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "midmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_tag: "2.1.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    - "scipy"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "scanpy"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+build_info:
+  config: "src/qc/calculate_qc_metrics/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker"
+  output: "target/nextflow/qc/calculate_qc_metrics"
+  executable: "target/nextflow/qc/calculate_qc_metrics/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "a0c9522486585774f76416150f8a3291409b5363"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  git_tag: "2.1.1-2-ga0c95224865"
+package_config:
+  name: "openpipeline"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".version := \"2.1.2\""
+  - ".engines[.type == 'docker'].target_tag := '2.1.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "openpipelines-bio"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/compress_h5mu.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/main.nf b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/main.nf
new file mode 100644
index 0000000..5fe7c0b
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/main.nf
@@ -0,0 +1,4218 @@
+// calculate_qc_metrics 2.1.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "calculate_qc_metrics",
+  "namespace" : "qc",
+  "version" : "2.1.2",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .obs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "example" : [
+            "ercc,highly_variable,mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--var_qc_metrics_fill_na_value",
+          "description" : "Fill any 'NA' values found in the columns specified with --var_qc_metrics to 'True' or 'False'.\nas False.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20;50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .var",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "The compression format to be used on the output h5mu object.",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "midmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_tag" : "2.1.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1",
+            "scipy"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scanpy"
+          ],
+          "upgrade" : true
+        }
+      ]
+    }
+  ],
+  "build_info" : {
+    "config" : "/home/runner/work/openpipeline/openpipeline/src/qc/calculate_qc_metrics/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker",
+    "output" : "/home/runner/work/openpipeline/openpipeline/target/nextflow/qc/calculate_qc_metrics",
+    "viash_version" : "0.9.4",
+    "git_commit" : "a0c9522486585774f76416150f8a3291409b5363",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline",
+    "git_tag" : "2.1.1-2-ga0c95224865"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/home/runner/work/openpipeline/openpipeline/src",
+    "target" : "/home/runner/work/openpipeline/openpipeline/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".version := \\"2.1.2\\"",
+      ".engines[.type == 'docker'].target_tag := '2.1.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "openpipelines-bio",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from mudata import read_h5ad
+from scipy.sparse import issparse, csr_array
+import numpy as np
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_qc_metrics': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'var_qc_metrics_fill_na_value': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'top_n_vars': $( if [ ! -z ${VIASH_PAR_TOP_N_VARS+x} ]; then echo "list(map(int, r'${VIASH_PAR_TOP_N_VARS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'output_obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_obs_total_counts_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_num_nonzero_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_total_counts_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_obs_mean': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_OBS_MEAN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_OBS_MEAN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_pct_dropout': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def count_nonzero(layer, axis):
+    """
+    This method is the functional equivalent of the old .getnnz function from scirpy,
+    but that function was deprecated. So we use the nonzero function to mimic the old
+    behavior.
+    """
+    axis ^= 1
+    nonzero_counts = dict(zip(*np.unique(layer.nonzero()[axis], return_counts=True)))
+    nonzero_per_axis_item = {
+        row_index: nonzero_counts.get(row_index, 0)
+        for row_index in range(layer.shape[axis])
+    }
+    return np.array(list(nonzero_per_axis_item.values()), dtype="int64")
+
+
+def main():
+    modality_data = read_h5ad(par["input"], mod=par["modality"])
+    var = modality_data.var
+    layer = modality_data.X if not par["layer"] else modality_data.layers[par["layer"]]
+    if not issparse(layer):
+        raise NotImplementedError("Expected layer to be in sparse format.")
+    layer = csr_array(layer)
+    layer.eliminate_zeros()
+
+    var_columns_to_add = {}
+
+    # var statistics
+    if par["output_var_obs_mean"]:
+        obs_mean = layer.mean(axis=0)
+        var_columns_to_add[par["output_var_obs_mean"]] = obs_mean
+    if par["output_var_total_counts_obs"]:
+        # from the np.sum documentation:
+        # Especially when summing a large number of lower precision floating point numbers,
+        # such as float32, numerical errors can become significant. In such cases it can
+        # be advisable to use dtype="float64" to use a higher precision for the output.
+        layer_with_type = layer
+        if np.issubdtype(layer.dtype, np.floating) and np.can_cast(
+            layer.dtype, np.float64, casting="safe"
+        ):
+            # 'safe' casting makes sure not to cast np.float128 or anything else to a lower precision dtype
+            layer_with_type = layer.astype(np.float64)
+        total_counts_obs = np.ravel(layer_with_type.sum(axis=0))
+        var_columns_to_add[par["output_var_total_counts_obs"]] = total_counts_obs
+
+    num_nonzero_obs = count_nonzero(layer, axis=0)
+    if par["output_var_num_nonzero_obs"]:
+        var_columns_to_add[par["output_var_num_nonzero_obs"]] = num_nonzero_obs
+    if par["output_var_pct_dropout"]:
+        var_columns_to_add[par["output_var_pct_dropout"]] = (
+            1 - num_nonzero_obs / layer.shape[0]
+        ) * 100
+
+    modality_data.var = modality_data.var.assign(**var_columns_to_add)
+
+    # obs statistics
+    obs_columns_to_add = {}
+    total_counts_var = np.ravel(layer.sum(axis=1))
+
+    if par["output_obs_num_nonzero_vars"]:
+        num_nonzero_vars = count_nonzero(layer, axis=1)
+        obs_columns_to_add[par["output_obs_num_nonzero_vars"]] = num_nonzero_vars
+
+    if par["output_obs_total_counts_vars"]:
+        obs_columns_to_add[par["output_obs_total_counts_vars"]] = total_counts_var
+
+    top_metrics = {}
+    if par["top_n_vars"]:
+        par["top_n_vars"] = sorted(par["top_n_vars"])
+        distributions = get_top_from_csr_matrix(layer, par["top_n_vars"])
+        top_metrics = {
+            distribution_size: distribution * 100
+            for distribution_size, distribution in zip(
+                par["top_n_vars"], distributions.T
+            )
+        }
+        obs_columns_to_add |= {
+            f"pct_of_counts_in_top_{n_top}_vars": col
+            for n_top, col in top_metrics.items()
+        }
+
+    if par["var_qc_metrics"]:
+        print(f"qc_metrics: {par['var_qc_metrics']}")
+        for qc_metric in par["var_qc_metrics"]:
+            if qc_metric not in var:
+                raise ValueError(
+                    f"Value for --var_qc_metrics, '{qc_metric}' "
+                    f"not found in .var for modality {par['modality']}"
+                )
+            qc_column = var[qc_metric]
+            if qc_column.isna().any():
+                if par["var_qc_metrics_fill_na_value"] is None:
+                    raise ValueError(
+                        f"The .var column '{qc_metric}', selected by '--var_qc_metrics', contains NA values. "
+                        "It is ambiguous whether or not to include these values in the static calulation. "
+                        "You can explicitly map the NA values to 'False' or 'True using '--var_qc_metrics_fill_na_value'"
+                    )
+                else:
+                    qc_column = qc_column.fillna(
+                        par["var_qc_metrics_fill_na_value"], inplace=False
+                    )
+            qc_column = qc_column.to_list()
+            if set(np.unique(qc_column)) - {True, False}:
+                raise ValueError(
+                    f"Column {qc_metric} in .var for modality {par['modality']} "
+                    f"must only contain boolean values"
+                )
+            total_counts_qc_metric = np.ravel(layer[:, qc_column].sum(axis=1))
+            obs_columns_to_add |= {
+                f"total_counts_{qc_metric}": total_counts_qc_metric,
+                f"pct_{qc_metric}": total_counts_qc_metric / total_counts_var * 100,
+            }
+
+    modality_data.obs = modality_data.obs.assign(**obs_columns_to_add)
+    write_h5ad_to_h5mu_with_compression(
+        par["output"],
+        par["input"],
+        par["modality"],
+        modality_data,
+        par["output_compression"],
+    )
+
+
+def get_top_from_csr_matrix(array, top_n_genes):
+    # csr matrices stores a 3D matrix in a format such that data for individual cells
+    # are stored in 1 array. Another array (indptr) here stores the ranges of indices
+    # to select from the data-array (.e.g. data[indptr[0]:indptr[1]] for row 0) for each row.
+    # Another array 'indices' maps each element of data to a column
+    # (data and indices arrays have the same length)
+    top_n_genes = np.array(top_n_genes).astype(np.int64)
+    assert np.all(top_n_genes[:-1] <= top_n_genes[1:]), "top_n_genes must be sorted"
+    row_indices, data = array.indptr, array.data
+    number_of_rows, max_genes_to_parse = row_indices.size - 1, top_n_genes[-1]
+    top_data = np.zeros((number_of_rows, max_genes_to_parse), dtype=data.dtype)
+    # Loop over each row to create a dense matrix without the 0 counts,
+    # but not for the whole matrix, only store the genes up until
+    # the largest number of top n genes.
+    for row_number in range(number_of_rows):
+        row_start_index, row_end_index = (
+            row_indices[row_number],
+            row_indices[row_number + 1],
+        )
+        row_data = data[row_start_index:row_end_index]  # all non-zero counts for an row
+        try:
+            # There are less genes with counts in the row than the
+            # maximum number of genes we would like to select
+            # all these genes are in the top genes, just store them
+            top_data[row_number, : row_end_index - row_start_index] = row_data
+        except ValueError:
+            # Store the counts for the top genes
+            top_data[row_number, :] = np.partition(row_data, -max_genes_to_parse)[
+                -max_genes_to_parse:
+            ]
+
+    # Partition works from smallest to largest, but we want largest
+    # so do smallest to largest first (but with reversed indices)
+    top_data = np.partition(top_data, max_genes_to_parse - top_n_genes)
+    # And then switch the order around
+    top_data = np.flip(top_data, axis=1)
+
+    cumulative = top_data.cumsum(axis=1, dtype=np.float64)[:, top_n_genes - 1]
+    return cumulative / np.expand_dims(array.sum(axis=1), 1)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "ghcr.io",
+    "image" : "openpipelines-bio/openpipeline/qc/calculate_qc_metrics",
+    "tag" : "2.1.0"
+  },
+  "label" : [
+    "singlecpu",
+    "midmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow.config
new file mode 100644
index 0000000..c1a67ba
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'qc/calculate_qc_metrics'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = '2.1.2'
+  description = 'Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_params.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_params.yaml
new file mode 100644
index 0000000..47537b9
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_params.yaml
@@ -0,0 +1,27 @@
+# Inputs
+input: # please fill in - example: "input.h5mu"
+modality: "rna"
+# layer: "raw_counts"
+
+# Metrics added to .obs
+# var_qc_metrics: ["ercc,highly_variable,mitochondrial"]
+# var_qc_metrics_fill_na_value: true
+# top_n_vars: [123]
+output_obs_num_nonzero_vars: "num_nonzero_vars"
+output_obs_total_counts_vars: "total_counts"
+
+# Metrics added to .var
+output_var_num_nonzero_obs: "num_nonzero_obs"
+output_var_total_counts_obs: "total_counts"
+output_var_obs_mean: "obs_mean"
+output_var_pct_dropout: "pct_dropout"
+
+# Outputs
+# output: "$id.$key.output.h5mu"
+# output_compression: "gzip"
+
+# Nextflow input-output arguments
+publish_dir: # please fill in - example: "output/"
+# param_list: "my_params.yaml"
+
+# Arguments
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_schema.json b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
new file mode 100644
index 0000000..d549dd9
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
@@ -0,0 +1,259 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "calculate_qc_metrics",
+"description": "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -\u003e name in scanpy):\n  - pct_dropout -\u003e pct_dropout_by_{expr_type}\n  - num_nonzero_obs -\u003e n_cells_by_{expr_type}\n  - obs_mean -\u003e mean_{expr_type}\n  - total_counts -\u003e total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -\u003e n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -\u003e pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -\u003e total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -\u003e pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -\u003e total_{expr_type}\n    \n",
+"type": "object",
+"definitions": {
+
+    
+    "Dataset input": {
+        "title": "Dataset input",
+        "type": "object",
+        "description": "Dataset input using nf-tower \"dataset\" or \"data explorer\". Allows for the input of multiple         parameter sets to initialise a Nextflow channel.",
+        "properties": {
+            "param_list": {
+                "description": "Dataset input can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml                 blob. The names of the input fields (e.g. csv columns, json keys) need to be an exact match with the workflow input parameters.",
+                "default": "",
+                "format": "file-path",
+                "mimetype": "text/csv",
+                "pattern": "^\\S+\\.csv$"
+            }
+        }
+    },
+
+    
+    
+    "inputs" : {
+    "title": "Inputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, example: `input.h5mu`. Input h5mu file",
+                "help_text": "Type: `file`, required, example: `input.h5mu`. Input h5mu file"
+            
+            }
+    
+
+        ,
+                "modality": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `rna`. ",
+                "help_text": "Type: `string`, default: `rna`. "
+            ,
+                "default":"rna"
+            }
+    
+
+        ,
+                "layer": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `raw_counts`. ",
+                "help_text": "Type: `string`, example: `raw_counts`. "
+            
+            }
+    
+
+}
+},
+    
+    
+    "outputs" : {
+    "title": "Outputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Output h5mu file",
+                "help_text": "Type: `file`, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Output h5mu file."
+            ,
+                "default":"$id.$key.output.h5mu"
+            }
+    
+
+        ,
+                "output_compression": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object",
+                "help_text": "Type: `string`, example: `gzip`, choices: ``gzip`, `lzf``. The compression format to be used on the output h5mu object.",
+                "enum": ["gzip", "lzf"]
+            
+            
+            }
+    
+
+}
+},
+    
+    
+    "metrics added to .obs" : {
+    "title": "Metrics added to .obs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "var_qc_metrics": {
+                "type":
+                "string",
+                "description": "Type: List of `string`, example: `ercc,highly_variable,mitochondrial`, multiple_sep: `\";\"`. Keys to select a boolean (containing only True or False) column from ",
+                "help_text": "Type: List of `string`, example: `ercc,highly_variable,mitochondrial`, multiple_sep: `\";\"`. Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled \u0027True\u0027, \ncompared to the total sum of the values for all genes.\n"
+            
+            }
+    
+
+        ,
+                "var_qc_metrics_fill_na_value": {
+                "type":
+                "boolean",
+                "description": "Type: `boolean`. Fill any \u0027NA\u0027 values found in the columns specified with --var_qc_metrics to \u0027True\u0027 or \u0027False\u0027",
+                "help_text": "Type: `boolean`. Fill any \u0027NA\u0027 values found in the columns specified with --var_qc_metrics to \u0027True\u0027 or \u0027False\u0027.\nas False.\n"
+            
+            }
+    
+
+        ,
+                "top_n_vars": {
+                "type":
+                "string",
+                "description": "Type: List of `integer`, multiple_sep: `\";\"`. Number of top vars to be used to calculate cumulative proportions",
+                "help_text": "Type: List of `integer`, multiple_sep: `\";\"`. Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20;50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n"
+            
+            }
+    
+
+        ,
+                "output_obs_num_nonzero_vars": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `num_nonzero_vars`. Name of column in ",
+                "help_text": "Type: `string`, default: `num_nonzero_vars`. Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n"
+            ,
+                "default":"num_nonzero_vars"
+            }
+    
+
+        ,
+                "output_obs_total_counts_vars": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `total_counts`. Name of the column for ",
+                "help_text": "Type: `string`, default: `total_counts`. Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n"
+            ,
+                "default":"total_counts"
+            }
+    
+
+}
+},
+    
+    
+    "metrics added to .var" : {
+    "title": "Metrics added to .var",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "output_var_num_nonzero_obs": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `num_nonzero_obs`. Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+                "help_text": "Type: `string`, default: `num_nonzero_obs`. Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n"
+            ,
+                "default":"num_nonzero_obs"
+            }
+    
+
+        ,
+                "output_var_total_counts_obs": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `total_counts`. Name of the column in ",
+                "help_text": "Type: `string`, default: `total_counts`. Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n"
+            ,
+                "default":"total_counts"
+            }
+    
+
+        ,
+                "output_var_obs_mean": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `obs_mean`. Name of the column in ",
+                "help_text": "Type: `string`, default: `obs_mean`. Name of the column in .obs providing the mean of the values in each row.\n"
+            ,
+                "default":"obs_mean"
+            }
+    
+
+        ,
+                "output_var_pct_dropout": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `pct_dropout`. Name of the column in ",
+                "help_text": "Type: `string`, default: `pct_dropout`. Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--num_nonzero_obs`\nbut percentage based.\n"
+            ,
+                "default":"pct_dropout"
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/inputs"
+    },
+
+    {
+    "$ref": "#/definitions/outputs"
+    },
+
+    {
+    "$ref": "#/definitions/metrics added to .obs"
+    },
+
+    {
+    "$ref": "#/definitions/metrics added to .var"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/setup_logger.py b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/qc/calculate_qc_metrics/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/.config.vsh.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/.config.vsh.yaml
new file mode 100644
index 0000000..2fa2c20
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/.config.vsh.yaml
@@ -0,0 +1,406 @@
+name: "qc"
+namespace: "workflows/qc"
+version: "2.1.2"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer to calculate qc metrics for."
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Mitochondrial & Ribosomal Gene Detection"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column name to be used to detect mitochondrial/ribosomal genes\
+      \ instead of .var_names (default if not set).\nGene names matching with the\
+      \ regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will\
+      \ be \nidentified as mitochondrial or ribosomal genes, respectively.\n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: ".Obs slot to store the fraction of reads found to be mitochondrial.\
+      \ Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--output_var_num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to add basic qc statistics to a MuData "
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "concat_test_data"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  test_dependencies:
+  - name: "qc_test"
+    namespace: "test_workflows/qc"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "metadata/grep_annotation_column"
+  repository:
+    type: "local"
+- name: "qc/calculate_qc_metrics"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+build_info:
+  config: "src/workflows/qc/qc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/qc/qc"
+  executable: "target/nextflow/workflows/qc/qc/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "a0c9522486585774f76416150f8a3291409b5363"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  git_tag: "2.1.1-2-ga0c95224865"
+  dependencies:
+  - "target/nextflow/metadata/grep_annotation_column"
+  - "target/nextflow/qc/calculate_qc_metrics"
+package_config:
+  name: "openpipeline"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".version := \"2.1.2\""
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "openpipelines-bio"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/main.nf b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/main.nf
new file mode 100644
index 0000000..6f095e8
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/main.nf
@@ -0,0 +1,3758 @@
+// qc 2.1.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "qc",
+  "namespace" : "workflows/qc",
+  "version" : "2.1.2",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer to calculate qc metrics for.",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Mitochondrial & Ribosomal Gene Detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : ".Obs slot to store the fraction of reads found to be mitochondrial. Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--output_var_num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to add basic qc statistics to a MuData ",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "qc_test",
+        "namespace" : "test_workflows/qc"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/grep_annotation_column",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "qc/calculate_qc_metrics",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/home/runner/work/openpipeline/openpipeline/src/workflows/qc/qc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/home/runner/work/openpipeline/openpipeline/target/nextflow/workflows/qc/qc",
+    "viash_version" : "0.9.4",
+    "git_commit" : "a0c9522486585774f76416150f8a3291409b5363",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline",
+    "git_tag" : "2.1.1-2-ga0c95224865"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/home/runner/work/openpipeline/openpipeline/src",
+    "target" : "/home/runner/work/openpipeline/openpipeline/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".version := \\"2.1.2\\""
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "openpipelines-bio",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { grep_annotation_column } from "${meta.resources_dir}/../../../../nextflow/metadata/grep_annotation_column/main.nf"
+include { calculate_qc_metrics } from "${meta.resources_dir}/../../../../nextflow/qc/calculate_qc_metrics/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+    output_ch = input_ch
+    // Avoid conflict between output from component and output for this workflow
+    | map {id, state -> 
+      assert state.output, "Output must be defined"
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // Add default for var_qc_metrics component
+    | map {id, state ->
+      def var_qc_default = []
+      // Remove the var_qc_metric argument from the state if its value is null (not specified)
+      def new_state = state.findAll { it.key != "var_qc_metrics" || it.value == null }
+      if (state.var_name_mitochondrial_genes) {
+        var_qc_default.add(state.var_name_mitochondrial_genes)
+      }
+      if (state.var_name_ribosomal_genes) {
+        var_qc_default.add(state.var_name_ribosomal_genes)
+      }
+      // Get the new state, but make sure to overwrite var_qc_metrics if the user has set it.
+      new_state = ["var_qc_metrics": var_qc_default.join(";")] + new_state
+      [id, new_state]
+    }
+
+    | grep_annotation_column.run(
+      key: "grep_mitochondrial_genes",
+      runIf: { id, state ->
+        state.var_name_mitochondrial_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_mitochondrial_genes,
+          "regex_pattern": state.mitochondrial_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_mitochondrial_fraction ? state.obs_name_mitochondrial_fraction: "fraction_$state.var_name_mitochondrial_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | grep_annotation_column.run(
+      key: "grep_ribosomal_genes",
+      runIf: { id, state ->
+        state.var_name_ribosomal_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_ribosomal_genes,
+          "regex_pattern": state.ribosomal_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_ribosomal_fraction ? state.obs_name_ribosomal_fraction: "fraction_$state.var_name_ribosomal_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | calculate_qc_metrics.run(
+      fromState: { id, state ->
+        def newState = [
+          "input": state.input,
+          "modality": state.modality,
+          "layer": state.layer,
+          // TODO: remove this workaround when Viash issue is resolved:
+          //       'top_n_vars': list(map(int, r''.split(';'))),
+          //     ValueError: invalid literal for int() with base 10: ''
+          // See https://github.com/viash-io/viash/issues/619
+          "top_n_vars": state.top_n_vars ? state.top_n_vars : null,
+          "var_qc_metrics_fill_na_value": state.var_qc_metrics_fill_na_value,
+          "output_obs_num_nonzero_vars": state.output_obs_num_nonzero_vars,
+          "output_obs_total_counts_vars": state.output_obs_total_counts_vars,
+          "output_var_num_nonzero_obs": state.output_var_num_nonzero_obs,
+          "output_var_total_counts_obs": state.output_var_total_counts_obs,
+          "output_var_obs_mean": state.output_var_obs_mean,
+          "output_var_pct_dropout": state.output_var_pct_dropout,
+          "output": state.workflow_output,
+          "compression": "gzip"
+        ]
+        if (state.var_qc_metrics) {
+          newState += ["var_qc_metrics": state.var_qc_metrics]
+        }
+      return newState
+      },
+      toState: ["output": "output"]
+    )
+    | setState(["output"]) 
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow.config
new file mode 100644
index 0000000..c073fb4
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/qc/qc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = '2.1.2'
+  description = 'A pipeline to add basic qc statistics to a MuData '
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_params.yaml b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_params.yaml
new file mode 100644
index 0000000..435f0c8
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_params.yaml
@@ -0,0 +1,33 @@
+# Inputs
+id: # please fill in - example: "foo"
+input: # please fill in - example: "input.h5mu"
+modality: "rna"
+# layer: "raw_counts"
+
+# Mitochondrial & Ribosomal Gene Detection
+# var_gene_names: "gene_symbol"
+# var_name_mitochondrial_genes: "foo"
+# obs_name_mitochondrial_fraction: "foo"
+mitochondrial_gene_regex: "^[mM][tT]-"
+# var_name_ribosomal_genes: "foo"
+# obs_name_ribosomal_fraction: "foo"
+ribosomal_gene_regex: "^[Mm]?[Rr][Pp][LlSs]"
+
+# QC metrics calculation options
+# var_qc_metrics: ["ercc,highly_variable"]
+top_n_vars: [50, 100, 200, 500]
+output_obs_num_nonzero_vars: "num_nonzero_vars"
+output_obs_total_counts_vars: "total_counts"
+output_var_num_nonzero_obs: "num_nonzero_obs"
+output_var_total_counts_obs: "total_counts"
+output_var_obs_mean: "obs_mean"
+output_var_pct_dropout: "pct_dropout"
+
+# Outputs
+# output: "$id.$key.output.h5mu"
+
+# Nextflow input-output arguments
+publish_dir: # please fill in - example: "output/"
+# param_list: "my_params.yaml"
+
+# Arguments
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_schema.json b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_schema.json
new file mode 100644
index 0000000..e8d9d26
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/nextflow_schema.json
@@ -0,0 +1,320 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "qc",
+"description": "A pipeline to add basic qc statistics to a MuData ",
+"type": "object",
+"definitions": {
+
+    
+    "Dataset input": {
+        "title": "Dataset input",
+        "type": "object",
+        "description": "Dataset input using nf-tower \"dataset\" or \"data explorer\". Allows for the input of multiple         parameter sets to initialise a Nextflow channel.",
+        "properties": {
+            "param_list": {
+                "description": "Dataset input can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml                 blob. The names of the input fields (e.g. csv columns, json keys) need to be an exact match with the workflow input parameters.",
+                "default": "",
+                "format": "file-path",
+                "mimetype": "text/csv",
+                "pattern": "^\\S+\\.csv$"
+            }
+        }
+    },
+
+    
+    
+    "inputs" : {
+    "title": "Inputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "id": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `foo`. ID of the sample",
+                "help_text": "Type: `string`, required, example: `foo`. ID of the sample."
+            
+            }
+    
+
+        ,
+                "input": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, example: `input.h5mu`. Path to the sample",
+                "help_text": "Type: `file`, required, example: `input.h5mu`. Path to the sample."
+            
+            }
+    
+
+        ,
+                "modality": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `rna`. Which modality to process",
+                "help_text": "Type: `string`, default: `rna`. Which modality to process."
+            ,
+                "default":"rna"
+            }
+    
+
+        ,
+                "layer": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `raw_counts`. Layer to calculate qc metrics for",
+                "help_text": "Type: `string`, example: `raw_counts`. Layer to calculate qc metrics for."
+            
+            }
+    
+
+}
+},
+    
+    
+    "outputs" : {
+    "title": "Outputs",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Destination path to the output",
+                "help_text": "Type: `file`, required, default: `$id.$key.output.h5mu`, example: `output.h5mu`. Destination path to the output."
+            ,
+                "default":"$id.$key.output.h5mu"
+            }
+    
+
+}
+},
+    
+    
+    "mitochondrial & ribosomal gene detection" : {
+    "title": "Mitochondrial & Ribosomal Gene Detection",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "var_gene_names": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `gene_symbol`. ",
+                "help_text": "Type: `string`, example: `gene_symbol`. .var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n"
+            
+            }
+    
+
+        ,
+                "var_name_mitochondrial_genes": {
+                "type":
+                "string",
+                "description": "Type: `string`. In which ",
+                "help_text": "Type: `string`. In which .var slot to store a boolean array corresponding the mitochondrial genes.\n"
+            
+            }
+    
+
+        ,
+                "obs_name_mitochondrial_fraction": {
+                "type":
+                "string",
+                "description": "Type: `string`. ",
+                "help_text": "Type: `string`. .Obs slot to store the fraction of reads found to be mitochondrial. Defaults to \u0027fraction_\u0027 suffixed by the value of --var_name_mitochondrial_genes\n"
+            
+            }
+    
+
+        ,
+                "mitochondrial_gene_regex": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `^[mM][tT]-`. Regex string that identifies mitochondrial genes from --var_gene_names",
+                "help_text": "Type: `string`, default: `^[mM][tT]-`. Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+            ,
+                "default":"^[mM][tT]-"
+            }
+    
+
+        ,
+                "var_name_ribosomal_genes": {
+                "type":
+                "string",
+                "description": "Type: `string`. In which ",
+                "help_text": "Type: `string`. In which .var slot to store a boolean array corresponding the ribosomal genes.\n"
+            
+            }
+    
+
+        ,
+                "obs_name_ribosomal_fraction": {
+                "type":
+                "string",
+                "description": "Type: `string`. When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an ",
+                "help_text": "Type: `string`. When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n"
+            
+            }
+    
+
+        ,
+                "ribosomal_gene_regex": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `^[Mm]?[Rr][Pp][LlSs]`. Regex string that identifies ribosomal genes from --var_gene_names",
+                "help_text": "Type: `string`, default: `^[Mm]?[Rr][Pp][LlSs]`. Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n"
+            ,
+                "default":"^[Mm]?[Rr][Pp][LlSs]"
+            }
+    
+
+}
+},
+    
+    
+    "qc metrics calculation options" : {
+    "title": "QC metrics calculation options",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "var_qc_metrics": {
+                "type":
+                "string",
+                "description": "Type: List of `string`, example: `ercc,highly_variable`, multiple_sep: `\",\"`. Keys to select a boolean (containing only True or False) column from ",
+                "help_text": "Type: List of `string`, example: `ercc,highly_variable`, multiple_sep: `\",\"`. Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled \u0027True\u0027, \ncompared to the total sum of the values for all genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n"
+            
+            }
+    
+
+        ,
+                "top_n_vars": {
+                "type":
+                "string",
+                "description": "Type: List of `integer`, default: `50,100,200,500`, multiple_sep: `\",\"`. Number of top vars to be used to calculate cumulative proportions",
+                "help_text": "Type: List of `integer`, default: `50,100,200,500`, multiple_sep: `\",\"`. Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n"
+            ,
+                "default":"50,100,200,500"
+            }
+    
+
+        ,
+                "output_obs_num_nonzero_vars": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `num_nonzero_vars`. Name of column in ",
+                "help_text": "Type: `string`, default: `num_nonzero_vars`. Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n"
+            ,
+                "default":"num_nonzero_vars"
+            }
+    
+
+        ,
+                "output_obs_total_counts_vars": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `total_counts`. Name of the column for ",
+                "help_text": "Type: `string`, default: `total_counts`. Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n"
+            ,
+                "default":"total_counts"
+            }
+    
+
+        ,
+                "output_var_num_nonzero_obs": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `num_nonzero_obs`. Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+                "help_text": "Type: `string`, default: `num_nonzero_obs`. Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n"
+            ,
+                "default":"num_nonzero_obs"
+            }
+    
+
+        ,
+                "output_var_total_counts_obs": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `total_counts`. Name of the column in ",
+                "help_text": "Type: `string`, default: `total_counts`. Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n"
+            ,
+                "default":"total_counts"
+            }
+    
+
+        ,
+                "output_var_obs_mean": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `obs_mean`. Name of the column in ",
+                "help_text": "Type: `string`, default: `obs_mean`. Name of the column in .obs providing the mean of the values in each row.\n"
+            ,
+                "default":"obs_mean"
+            }
+    
+
+        ,
+                "output_var_pct_dropout": {
+                "type":
+                "string",
+                "description": "Type: `string`, default: `pct_dropout`. Name of the column in ",
+                "help_text": "Type: `string`, default: `pct_dropout`. Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--output_var_num_nonzero_obs`\nbut percentage based.\n"
+            ,
+                "default":"pct_dropout"
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/inputs"
+    },
+
+    {
+    "$ref": "#/definitions/outputs"
+    },
+
+    {
+    "$ref": "#/definitions/mitochondrial & ribosomal gene detection"
+    },
+
+    {
+    "$ref": "#/definitions/qc metrics calculation options"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/integration_tests.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels.config
new file mode 100644
index 0000000..8aca1a0
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 25.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 50.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels_ci.config b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml
new file mode 100644
index 0000000..8a58ac6
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml
@@ -0,0 +1,187 @@
+name: "move_files_to_directory"
+version: "v0.2.0"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "maintainer"
+  info:
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Paths of the files that will be copied into the output directory."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "Path to output directory"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+summary: "Publish one or multiple files to the same directory"
+description: "This component copies one or multiple files to the same destination\
+  \ directory, creating the output directory if it doesn't exist."
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+license: "MIT"
+links:
+  repository: "https://github.com/viash-hub/craftbox"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "debian:latest"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.2.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/move_files_to_directory/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/move_files_to_directory"
+  executable: "target/nextflow/move_files_to_directory/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "1c1b0a4a1aff891ab678072b0ba915ac3ac71610"
+  git_remote: "https://github.com/viash-hub/craftbox"
+  git_tag: "v0.1.0-8-g1c1b0a4"
+package_config:
+  name: "craftbox"
+  version: "v0.2.0"
+  summary: "A collection of custom-tailored scripts and applied utilities built with\
+    \ Viash.\n"
+  description: "`craftbox` is a curated collection of custom scripts and utilities\
+    \ designed to tackle context-specific tasks.\n\nEmphasizing the Viash principles,\
+    \ `craftbox` components aim for **reusability**, **reproducibility**, and adherence\
+    \ to **best practices**. Key features generally include:\n\n* **Standalone & Nextflow\
+    \ Ready:** Components are built to run directly via the command line or be smoothly\
+    \ integrated into Nextflow workflows.\n* **Custom Implementations:** Contains\
+    \ scripts and tools developed for particular tasks that may not be found in broader\
+    \ collections.\n* **High Quality Standards (promoted by Viash):**\n    * Clear\
+    \ documentation for components and their parameters.\n    * Full exposure of underlying\
+    \ script/tool arguments for fine-grained control.\n    * Containerized (Docker)\
+    \ to ensure dependency management and a consistent, reproducible runtime environment.\n\
+    \    * Unit tested where applicable to ensure components function as expected.\n"
+  info: null
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.2.0'"
+  keywords:
+  - "scripts"
+  - "custom"
+  - "implementations"
+  - "utilities"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/viash-hub/craftbox"
+    issue_tracker: "https://github.com/viash-hub/craftbox/issues"
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf
new file mode 100644
index 0000000..f4e3ca1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf
@@ -0,0 +1,3811 @@
+// move_files_to_directory v0.2.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "move_files_to_directory",
+  "version" : "v0.2.0",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Paths of the files that will be copied into the output directory.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Path to output directory",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    }
+  ],
+  "summary" : "Publish one or multiple files to the same directory",
+  "description" : "This component copies one or multiple files to the same destination directory, creating the output directory if it doesn't exist.",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/viash-hub/craftbox"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        }
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "debian:latest",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.2.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/move_files_to_directory/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "target/nextflow/move_files_to_directory",
+    "viash_version" : "0.9.4",
+    "git_commit" : "1c1b0a4a1aff891ab678072b0ba915ac3ac71610",
+    "git_remote" : "https://github.com/viash-hub/craftbox",
+    "git_tag" : "v0.1.0-8-g1c1b0a4"
+  },
+  "package_config" : {
+    "name" : "craftbox",
+    "version" : "v0.2.0",
+    "summary" : "A collection of custom-tailored scripts and applied utilities built with Viash.\n",
+    "description" : "`craftbox` is a curated collection of custom scripts and utilities designed to tackle context-specific tasks.\n\nEmphasizing the Viash principles, `craftbox` components aim for **reusability**, **reproducibility**, and adherence to **best practices**. Key features generally include:\n\n* **Standalone & Nextflow Ready:** Components are built to run directly via the command line or be smoothly integrated into Nextflow workflows.\n* **Custom Implementations:** Contains scripts and tools developed for particular tasks that may not be found in broader collections.\n* **High Quality Standards (promoted by Viash):**\n    * Clear documentation for components and their parameters.\n    * Full exposure of underlying script/tool arguments for fine-grained control.\n    * Containerized (Docker) to ensure dependency management and a consistent, reproducible runtime environment.\n    * Unit tested where applicable to ensure components function as expected.\n",
+    "viash_version" : "0.9.4",
+    "source" : "src",
+    "target" : "target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.2.0'"
+    ],
+    "keywords" : [
+      "scripts",
+      "custom",
+      "implementations",
+      "utilities"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/viash-hub/craftbox",
+      "issue_tracker" : "https://github.com/viash-hub/craftbox/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+#!/bin/bash
+
+set -eo pipefail
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "${VIASH_PAR_INPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input='&'#" ; else echo "# par_input="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "${VIASH_PAR_OUTPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output='&'#" ; else echo "# par_output="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+
+
+if [[ ! -d "\\$par_output" ]]; then
+  mkdir -p "\\$par_output"
+fi
+
+# Process multiple input files
+IFS=";" read -ra input_files <<< "\\$par_input"
+for file in "\\${input_files[@]}"; do
+  # Check if the file exists before copying
+  if [[ -f "\\$file" ]]; then
+    cp "\\$file" "\\$par_output/"
+    echo "Copied \\$file to \\$par_output/"
+  else
+    echo "Warning: Input file \\$file does not exist, skipping"
+  fi
+done
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/craftbox/move_files_to_directory",
+    "tag" : "v0.2.0"
+  },
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config
new file mode 100644
index 0000000..38bce9a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'move_files_to_directory'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.2.0'
+  description = 'This component copies one or multiple files to the same destination directory, creating the output directory if it doesn\'t exist.'
+  author = 'Dorien Roosen'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json
new file mode 100644
index 0000000..ddc6758
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json
@@ -0,0 +1,81 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "move_files_to_directory",
+"description": "This component copies one or multiple files to the same destination directory, creating the output directory if it doesn\u0027t exist.",
+"type": "object",
+"definitions": {
+
+    
+    
+    "arguments" : {
+    "title": "Arguments",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: List of `file`, required, multiple_sep: `\";\"`. Paths of the files that will be copied into the output directory",
+                "help_text": "Type: List of `file`, required, multiple_sep: `\";\"`. Paths of the files that will be copied into the output directory."
+            
+            }
+    
+
+        ,
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, default: `$id.$key.output`. Path to output directory",
+                "help_text": "Type: `file`, required, default: `$id.$key.output`. Path to output directory"
+            ,
+                "default":"$id.$key.output"
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        ,
+                "param_list": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `my_params.yaml`. Allows inputting multiple parameter sets to initialise a Nextflow channel",
+                "help_text": "Type: `string`, example: `my_params.yaml`. Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.\n\n* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ [\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027], [\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027] ]`.\n* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.\n* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027}, {\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027} ]`.\n* A yaml blob can also be passed directly as a string. Example: `--param_list \"[ {\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027}, {\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027} ]\"`.\n\nWhen passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.",
+                "hidden": true
+            
+            }
+    
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/arguments"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/executable/ingestion_qc/detect_ingestion_method/.config.vsh.yaml b/target/executable/ingestion_qc/detect_ingestion_method/.config.vsh.yaml
new file mode 100644
index 0000000..fac5f77
--- /dev/null
+++ b/target/executable/ingestion_qc/detect_ingestion_method/.config.vsh.yaml
@@ -0,0 +1,265 @@
+name: "detect_ingestion_method"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu file(s)"
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "The modality to use"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "string"
+    name: "--output_uns_ingestion_method"
+    description: "The .uns field in which to store the exprimental setup. Values stored\
+      \ are `cellranger_multi`, `xenium` or `cosmx`."
+    info: null
+    default:
+    - "ingestion_method"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "The output h5mu file, containing an .uns field with experiment description."
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Detects the ingestion method of a dataset.\nCurrently detects either\
+  \ 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other\
+  \ technologies upon request.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "sample_one.qc.h5mu"
+- type: "file"
+  path: "Lung5_Rep2_tiny.qc.h5mu"
+- type: "file"
+  path: "xenium_tiny.qc.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/detect_ingestion_method/config.vsh.yaml"
+  runner: "executable"
+  engine: "docker|native"
+  output: "target/executable/ingestion_qc/detect_ingestion_method"
+  executable: "target/executable/ingestion_qc/detect_ingestion_method/detect_ingestion_method"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/executable/ingestion_qc/detect_ingestion_method/detect_ingestion_method b/target/executable/ingestion_qc/detect_ingestion_method/detect_ingestion_method
new file mode 100755
index 0000000..04395e8
--- /dev/null
+++ b/target/executable/ingestion_qc/detect_ingestion_method/detect_ingestion_method
@@ -0,0 +1,1219 @@
+#!/usr/bin/env bash
+
+# detect_ingestion_method build_main
+# 
+# This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+# work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+# Intuitive.
+# 
+# The component may contain files which fall under a different license. The
+# authors of this component should specify the license in the header of such
+# files, or include a separate license file detailing the licenses of all included
+# files.
+# 
+# Component authors:
+#  * Dorien Roosen (author)
+#  * Weiwei Schultz (contributor)
+
+set -e
+
+if [ -z "$VIASH_TEMP" ]; then
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMP}
+  VIASH_TEMP=${VIASH_TEMP:-/tmp}
+fi
+
+# define helper functions
+# ViashQuote: put quotes around non flag values
+# $1     : unquoted string
+# return : possibly quoted string
+# examples:
+#   ViashQuote --foo      # returns --foo
+#   ViashQuote bar        # returns 'bar'
+#   Viashquote --foo=bar  # returns --foo='bar'
+function ViashQuote {
+  if [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+=.+$ ]]; then
+    echo "$1" | sed "s#=\(.*\)#='\1'#"
+  elif [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+$ ]]; then
+    echo "$1"
+  else
+    echo "'$1'"
+  fi
+}
+# ViashRemoveFlags: Remove leading flag
+# $1     : string with a possible leading flag
+# return : string without possible leading flag
+# examples:
+#   ViashRemoveFlags --foo=bar  # returns bar
+function ViashRemoveFlags {
+  echo "$1" | sed 's/^--*[a-zA-Z0-9_\-]*=//'
+}
+# ViashSourceDir: return the path of a bash file, following symlinks
+# usage   : ViashSourceDir ${BASH_SOURCE[0]}
+# $1      : Should always be set to ${BASH_SOURCE[0]}
+# returns : The absolute path of the bash file
+function ViashSourceDir {
+  local source="$1"
+  while [ -h "$source" ]; do
+    local dir="$( cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd )"
+    source="$(readlink "$source")"
+    [[ $source != /* ]] && source="$dir/$source"
+  done
+  cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd
+}
+# ViashFindTargetDir: return the path of the '.build.yaml' file, following symlinks
+# usage   : ViashFindTargetDir 'ScriptPath'
+# $1      : The location from where to start the upward search
+# returns : The absolute path of the '.build.yaml' file
+function ViashFindTargetDir {
+  local source="$1"
+  while [[ "$source" != "" && ! -e "$source/.build.yaml" ]]; do
+    source=${source%/*}
+  done
+  echo $source
+}
+# see https://en.wikipedia.org/wiki/Syslog#Severity_level
+VIASH_LOGCODE_EMERGENCY=0
+VIASH_LOGCODE_ALERT=1
+VIASH_LOGCODE_CRITICAL=2
+VIASH_LOGCODE_ERROR=3
+VIASH_LOGCODE_WARNING=4
+VIASH_LOGCODE_NOTICE=5
+VIASH_LOGCODE_INFO=6
+VIASH_LOGCODE_DEBUG=7
+VIASH_VERBOSITY=$VIASH_LOGCODE_NOTICE
+
+# ViashLog: Log events depending on the verbosity level
+# usage: ViashLog 1 alert Oh no something went wrong!
+# $1: required verbosity level
+# $2: display tag
+# $3+: messages to display
+# stdout: Your input, prepended by '[$2] '.
+function ViashLog {
+  local required_level="$1"
+  local display_tag="$2"
+  shift 2
+  if [ $VIASH_VERBOSITY -ge $required_level ]; then
+    >&2 echo "[$display_tag]" "$@"
+  fi
+}
+
+# ViashEmergency: log events when the system is unstable
+# usage: ViashEmergency Oh no something went wrong.
+# stdout: Your input, prepended by '[emergency] '.
+function ViashEmergency {
+  ViashLog $VIASH_LOGCODE_EMERGENCY emergency "$@"
+}
+
+# ViashAlert: log events when actions must be taken immediately (e.g. corrupted system database)
+# usage: ViashAlert Oh no something went wrong.
+# stdout: Your input, prepended by '[alert] '.
+function ViashAlert {
+  ViashLog $VIASH_LOGCODE_ALERT alert "$@"
+}
+
+# ViashCritical: log events when a critical condition occurs
+# usage: ViashCritical Oh no something went wrong.
+# stdout: Your input, prepended by '[critical] '.
+function ViashCritical {
+  ViashLog $VIASH_LOGCODE_CRITICAL critical "$@"
+}
+
+# ViashError: log events when an error condition occurs
+# usage: ViashError Oh no something went wrong.
+# stdout: Your input, prepended by '[error] '.
+function ViashError {
+  ViashLog $VIASH_LOGCODE_ERROR error "$@"
+}
+
+# ViashWarning: log potentially abnormal events
+# usage: ViashWarning Something may have gone wrong.
+# stdout: Your input, prepended by '[warning] '.
+function ViashWarning {
+  ViashLog $VIASH_LOGCODE_WARNING warning "$@"
+}
+
+# ViashNotice: log significant but normal events
+# usage: ViashNotice This just happened.
+# stdout: Your input, prepended by '[notice] '.
+function ViashNotice {
+  ViashLog $VIASH_LOGCODE_NOTICE notice "$@"
+}
+
+# ViashInfo: log normal events
+# usage: ViashInfo This just happened.
+# stdout: Your input, prepended by '[info] '.
+function ViashInfo {
+  ViashLog $VIASH_LOGCODE_INFO info "$@"
+}
+
+# ViashDebug: log all events, for debugging purposes
+# usage: ViashDebug This just happened.
+# stdout: Your input, prepended by '[debug] '.
+function ViashDebug {
+  ViashLog $VIASH_LOGCODE_DEBUG debug "$@"
+}
+
+# find source folder of this component
+VIASH_META_RESOURCES_DIR=`ViashSourceDir ${BASH_SOURCE[0]}`
+
+# find the root of the built components & dependencies
+VIASH_TARGET_DIR=`ViashFindTargetDir $VIASH_META_RESOURCES_DIR`
+
+# define meta fields
+VIASH_META_NAME="detect_ingestion_method"
+VIASH_META_FUNCTIONALITY_NAME="detect_ingestion_method"
+VIASH_META_EXECUTABLE="$VIASH_META_RESOURCES_DIR/$VIASH_META_NAME"
+VIASH_META_CONFIG="$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+VIASH_META_TEMP_DIR="$VIASH_TEMP"
+
+
+
+# initialise variables
+VIASH_MODE='run'
+VIASH_ENGINE_ID='docker'
+
+######## Helper functions for setting up Docker images for viash ########
+# expects: ViashDockerBuild
+
+# ViashDockerInstallationCheck: check whether Docker is installed correctly
+#
+# examples:
+#   ViashDockerInstallationCheck
+function ViashDockerInstallationCheck {
+  ViashDebug "Checking whether Docker is installed"
+  if [ ! command -v docker &> /dev/null ]; then
+    ViashCritical "Docker doesn't seem to be installed. See 'https://docs.docker.com/get-docker/' for instructions."
+    exit 1
+  fi
+
+  ViashDebug "Checking whether the Docker daemon is running"
+  local save=$-; set +e
+  local docker_version=$(docker version --format '{{.Client.APIVersion}}' 2> /dev/null)
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashCritical "Docker daemon does not seem to be running. Try one of the following:"
+    ViashCritical "- Try running 'dockerd' in the command line"
+    ViashCritical "- See https://docs.docker.com/config/daemon/"
+    exit 1
+  fi
+}
+
+# ViashDockerRemoteTagCheck: check whether a Docker image is available 
+# on a remote. Assumes `docker login` has been performed, if relevant.
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerRemoteTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerRemoteTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerRemoteTagCheck {
+  docker manifest inspect $1 > /dev/null 2> /dev/null
+}
+
+# ViashDockerLocalTagCheck: check whether a Docker image is available locally
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   docker pull python:latest
+#   ViashDockerLocalTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerLocalTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerLocalTagCheck {
+  [ -n "$(docker images -q $1)" ]
+}
+
+# ViashDockerPull: pull a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPull python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPull sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPull {
+  ViashNotice "Checking if Docker image is available at '$1'"
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker pull $1 && return 0 || return 1
+  else
+    local save=$-; set +e
+    docker pull $1 2> /dev/null > /dev/null
+    local out=$?
+    [[ $save =~ e ]] && set -e
+    if [ $out -ne 0 ]; then
+      ViashWarning "Could not pull from '$1'. Docker image doesn't exist or is not accessible."
+    fi
+    return $out
+  fi
+}
+
+# ViashDockerPush: push a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPush python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPush sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPush {
+  ViashNotice "Pushing image to '$1'"
+  local save=$-; set +e
+  local out
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker push $1
+    out=$?
+  else
+    docker push $1 2> /dev/null > /dev/null
+    out=$?
+  fi
+  [[ $save =~ e ]] && set -e
+  if [ $out -eq 0 ]; then
+    ViashNotice "Container '$1' push succeeded."
+  else
+    ViashError "Container '$1' push errored. You might not be logged in or have the necessary permissions."
+  fi
+  return $out
+}
+
+# ViashDockerPullElseBuild: pull a Docker image, else build it
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# ViashDockerBuild    : a Bash function which builds a docker image, takes image identifier as argument.
+# examples:
+#   ViashDockerPullElseBuild mynewcomponent
+function ViashDockerPullElseBuild {
+  local save=$-; set +e
+  ViashDockerPull $1
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashDockerBuild $@
+  fi
+}
+
+# ViashDockerSetup: create a Docker image, according to specified docker setup strategy
+#
+# $1          : image identifier with format `[registry/]image[:tag]`
+# $2          : docker setup strategy, see DockerSetupStrategy.scala
+# examples:
+#   ViashDockerSetup mynewcomponent alwaysbuild
+function ViashDockerSetup {
+  local image_id="$1"
+  local setup_strategy="$2"
+  if [ "$setup_strategy" == "alwaysbuild" -o "$setup_strategy" == "build" -o "$setup_strategy" == "b" ]; then
+    ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspull" -o "$setup_strategy" == "pull" -o "$setup_strategy" == "p" ]; then
+    ViashDockerPull $image_id
+  elif [ "$setup_strategy" == "alwayspullelsebuild" -o "$setup_strategy" == "pullelsebuild" ]; then
+    ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspullelsecachedbuild" -o "$setup_strategy" == "pullelsecachedbuild" ]; then
+    ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayscachedbuild" -o "$setup_strategy" == "cachedbuild" -o "$setup_strategy" == "cb" ]; then
+    ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [[ "$setup_strategy" =~ ^ifneedbe ]]; then
+    local save=$-; set +e
+    ViashDockerLocalTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashInfo "Image $image_id already exists"
+    elif [ "$setup_strategy" == "ifneedbebuild" ]; then
+      ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbecachedbuild" ]; then
+      ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepull" ]; then
+      ViashDockerPull $image_id
+    elif [ "$setup_strategy" == "ifneedbepullelsebuild" ]; then
+      ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepullelsecachedbuild" ]; then
+      ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    else
+      ViashError "Unrecognised Docker strategy: $setup_strategy"
+      exit 1
+    fi
+  elif [ "$setup_strategy" == "push" -o "$setup_strategy" == "forcepush" -o "$setup_strategy" == "alwayspush" ]; then
+    ViashDockerPush "$image_id"
+  elif [ "$setup_strategy" == "pushifnotpresent" -o "$setup_strategy" == "gentlepush" -o "$setup_strategy" == "maybepush" ]; then
+    local save=$-; set +e
+    ViashDockerRemoteTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashNotice "Container '$image_id' exists, doing nothing."
+    else
+      ViashNotice "Container '$image_id' does not yet exist."
+      ViashDockerPush "$image_id"
+    fi
+  elif [ "$setup_strategy" == "donothing" -o "$setup_strategy" == "meh" ]; then
+    ViashNotice "Skipping setup."
+  else
+    ViashError "Unrecognised Docker strategy: $setup_strategy"
+    exit 1
+  fi
+}
+
+# ViashDockerCheckCommands: Check whether a docker container has the required commands
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# $@                  : commands to verify being present
+# examples:
+#   ViashDockerCheckCommands bash:4.0 bash ps foo
+function ViashDockerCheckCommands {
+  local image_id="$1"
+  shift 1
+  local commands="$@"
+  local save=$-; set +e
+  local missing # mark 'missing' as local in advance, otherwise the exit code of the command will be missing and always be '0'
+  missing=$(docker run --rm --entrypoint=sh "$image_id" -c "for command in $commands; do command -v \$command >/dev/null 2>&1; if [ \$? -ne 0 ]; then echo \$command; exit 1; fi; done")
+  local outCheck=$?
+  [[ $save =~ e ]] && set -e
+  if [ $outCheck -ne 0 ]; then
+  	ViashError "Docker container '$image_id' does not contain command '$missing'."
+  	exit 1
+  fi
+}
+
+# ViashDockerBuild: build a docker image
+# $1                               : image identifier with format `[registry/]image[:tag]`
+# $...                             : additional arguments to pass to docker build
+# $VIASH_META_TEMP_DIR             : temporary directory to store dockerfile & optional resources in
+# $VIASH_META_NAME                 : name of the component
+# $VIASH_META_RESOURCES_DIR        : directory containing the resources
+# $VIASH_VERBOSITY                 : verbosity level
+# exit code $?                     : whether or not the image was built successfully
+function ViashDockerBuild {
+  local image_id="$1"
+  shift 1
+
+  # create temporary directory to store dockerfile & optional resources in
+  local tmpdir=$(mktemp -d "$VIASH_META_TEMP_DIR/dockerbuild-$VIASH_META_NAME-XXXXXX")
+  local dockerfile="$tmpdir/Dockerfile"
+  function clean_up {
+    rm -rf "$tmpdir"
+  }
+  trap clean_up EXIT
+
+  # store dockerfile and resources
+  ViashDockerfile "$VIASH_ENGINE_ID" > "$dockerfile"
+
+  # generate the build command
+  local docker_build_cmd="docker build -t '$image_id' $@ '$VIASH_META_RESOURCES_DIR' -f '$dockerfile'"
+
+  # build the container
+  ViashNotice "Building container '$image_id' with Dockerfile"
+  ViashInfo "$docker_build_cmd"
+  local save=$-; set +e
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    eval $docker_build_cmd
+  else
+    eval $docker_build_cmd &> "$tmpdir/docker_build.log"
+  fi
+
+  # check exit code
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashError "Error occurred while building container '$image_id'"
+    if [ $VIASH_VERBOSITY -lt $VIASH_LOGCODE_INFO ]; then
+      ViashError "Transcript: --------------------------------"
+      cat "$tmpdir/docker_build.log"
+      ViashError "End of transcript --------------------------"
+    fi
+    exit 1
+  fi
+}
+
+######## End of helper functions for setting up Docker images for viash ########
+
+# ViashDockerFile: print the dockerfile to stdout
+# $1    : engine identifier
+# return : dockerfile required to run this component
+# examples:
+#   ViashDockerFile
+function ViashDockerfile {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    cat << 'VIASHDOCKER'
+FROM python:3.12-slim
+ENTRYPOINT []
+RUN apt-get update && \
+  DEBIAN_FRONTEND=noninteractive apt-get install -y procps && \
+  rm -rf /var/lib/apt/lists/*
+
+RUN pip install --upgrade pip && \
+  pip install --upgrade --no-cache-dir "anndata~=0.11.1" "mudata~=0.3.1" && \
+  python -c 'exec("try:\n  import awkward\nexcept ModuleNotFoundError:\n  exit(0)\nelse:  exit(1)")'
+
+LABEL org.opencontainers.image.authors="Dorien Roosen, Weiwei Schultz"
+LABEL org.opencontainers.image.description="Companion container for running component ingestion_qc detect_ingestion_method"
+LABEL org.opencontainers.image.created="2025-08-11T08:32:06Z"
+LABEL org.opencontainers.image.source="https://github.com/openpipelines-bio/openpipeline_incubator"
+LABEL org.opencontainers.image.revision="dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+LABEL org.opencontainers.image.version="build_main"
+
+VIASHDOCKER
+  fi
+}
+
+# ViashDockerBuildArgs: return the arguments to pass to docker build
+# $1    : engine identifier
+# return : arguments to pass to docker build
+function ViashDockerBuildArgs {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    echo ""
+  fi
+}
+
+# ViashAbsolutePath: generate absolute path from relative path
+# borrowed from https://stackoverflow.com/a/21951256
+# $1     : relative filename
+# return : absolute path
+# examples:
+#   ViashAbsolutePath some_file.txt   # returns /path/to/some_file.txt
+#   ViashAbsolutePath /foo/bar/..     # returns /foo
+function ViashAbsolutePath {
+  local thePath
+  local parr
+  local outp
+  local len
+  if [[ ! "$1" =~ ^/ ]]; then
+    thePath="$PWD/$1"
+  else
+    thePath="$1"
+  fi
+  echo "$thePath" | (
+    IFS=/
+    read -a parr
+    declare -a outp
+    for i in "${parr[@]}"; do
+      case "$i" in
+      ''|.) continue ;;
+      ..)
+        len=${#outp[@]}
+        if ((len==0)); then
+          continue
+        else
+          unset outp[$((len-1))]
+        fi
+        ;;
+      *)
+        len=${#outp[@]}
+        outp[$len]="$i"
+      ;;
+      esac
+    done
+    echo /"${outp[*]}"
+  )
+}
+# ViashDockerAutodetectMount: auto configuring docker mounts from parameters
+# $1                             : The parameter value
+# returns                        : New parameter
+# $VIASH_DIRECTORY_MOUNTS        : Added another parameter to be passed to docker
+# $VIASH_DOCKER_AUTOMOUNT_PREFIX : The prefix to be used for the automounts
+# examples:
+#   ViashDockerAutodetectMount /path/to/bar      # returns '/viash_automount/path/to/bar'
+#   ViashDockerAutodetectMountArg /path/to/bar   # returns '--volume="/path/to:/viash_automount/path/to"'
+function ViashDockerAutodetectMount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  if [ -z "$base_name" ]; then
+    echo "$mount_target"
+  else
+    echo "$mount_target/$base_name"
+  fi
+}
+function ViashDockerAutodetectMountArg {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  ViashDebug "ViashDockerAutodetectMountArg $1 -> $mount_source -> $mount_target"
+  echo "--volume=\"$mount_source:$mount_target\""
+}
+function ViashDockerStripAutomount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  echo "${abs_path#$VIASH_DOCKER_AUTOMOUNT_PREFIX}"
+}
+# initialise variables
+VIASH_DIRECTORY_MOUNTS=()
+
+# configure default docker automount prefix if it is unset
+if [ -z "${VIASH_DOCKER_AUTOMOUNT_PREFIX+x}" ]; then
+  VIASH_DOCKER_AUTOMOUNT_PREFIX="/viash_automount"
+fi
+
+# initialise docker variables
+VIASH_DOCKER_RUN_ARGS=(-i --rm)
+
+
+# ViashHelp: Display helpful explanation about this executable
+function ViashHelp {
+  echo "detect_ingestion_method build_main"
+  echo ""
+  echo "Detects the ingestion method of a dataset."
+  echo "Currently detects either 10X CellRanger Multi, 10X Xenium or Nanostring CosMx,"
+  echo "but can be extended to other technologies upon request."
+  echo ""
+  echo "Inputs:"
+  echo "    --input"
+  echo "        type: file, required parameter, file must exist"
+  echo "        example: path/to/file.h5mu"
+  echo "        The input h5mu file(s)"
+  echo ""
+  echo "    --modality"
+  echo "        type: string"
+  echo "        default: rna"
+  echo "        The modality to use"
+  echo ""
+  echo "Outputs:"
+  echo "    --output_uns_ingestion_method"
+  echo "        type: string"
+  echo "        default: ingestion_method"
+  echo "        The .uns field in which to store the exprimental setup. Values stored"
+  echo "        are \`cellranger_multi\`, \`xenium\` or \`cosmx\`."
+  echo ""
+  echo "    --output"
+  echo "        type: file, required parameter, output, file must exist"
+  echo "        example: path/to/file.h5mu"
+  echo "        The output h5mu file, containing an .uns field with experiment"
+  echo "        description."
+  echo ""
+  echo "Viash built in Computational Requirements:"
+  echo "    ---cpus=INT"
+  echo "        Number of CPUs to use"
+  echo "    ---memory=STRING"
+  echo "        Amount of memory to use. Examples: 4GB, 3MiB."
+  echo ""
+  echo "Viash built in Docker:"
+  echo "    ---setup=STRATEGY"
+  echo "        Setup the docker container. Options are: alwaysbuild, alwayscachedbuild, ifneedbebuild, ifneedbecachedbuild, alwayspull, alwayspullelsebuild, alwayspullelsecachedbuild, ifneedbepull, ifneedbepullelsebuild, ifneedbepullelsecachedbuild, push, pushifnotpresent, donothing."
+  echo "        Default: ifneedbepullelsecachedbuild"
+  echo "    ---dockerfile"
+  echo "        Print the dockerfile to stdout."
+  echo "    ---docker_run_args=ARG"
+  echo "        Provide runtime arguments to Docker. See the documentation on \`docker run\` for more information."
+  echo "    ---docker_image_id"
+  echo "        Print the docker image id to stdout."
+  echo "    ---debug"
+  echo "        Enter the docker container for debugging purposes."
+  echo ""
+  echo "Viash built in Engines:"
+  echo "    ---engine=ENGINE_ID"
+  echo "        Specify the engine to use. Options are: docker, native."
+  echo "        Default: docker"
+}
+
+# initialise array
+VIASH_POSITIONAL_ARGS=''
+
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        -h|--help)
+            ViashHelp
+            exit
+            ;;
+        ---v|---verbose)
+            let "VIASH_VERBOSITY=VIASH_VERBOSITY+1"
+            shift 1
+            ;;
+        ---verbosity)
+            VIASH_VERBOSITY="$2"
+            shift 2
+            ;;
+        ---verbosity=*)
+            VIASH_VERBOSITY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        --version)
+            echo "detect_ingestion_method build_main"
+            exit
+            ;;
+        --input)
+            [ -n "$VIASH_PAR_INPUT" ] && ViashError Bad arguments for option \'--input\': \'$VIASH_PAR_INPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --input. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --input=*)
+            [ -n "$VIASH_PAR_INPUT" ] && ViashError Bad arguments for option \'--input=*\': \'$VIASH_PAR_INPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --modality)
+            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MODALITY="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --modality. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --modality=*)
+            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality=*\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_MODALITY=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --output_uns_ingestion_method)
+            [ -n "$VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD" ] && ViashError Bad arguments for option \'--output_uns_ingestion_method\': \'$VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_uns_ingestion_method. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --output_uns_ingestion_method=*)
+            [ -n "$VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD" ] && ViashError Bad arguments for option \'--output_uns_ingestion_method=*\': \'$VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --output)
+            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --output=*)
+            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output=*\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---engine)
+            VIASH_ENGINE_ID="$2"
+            shift 2
+            ;;
+        ---engine=*)
+            VIASH_ENGINE_ID="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---setup)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$2"
+            shift 2
+            ;;
+        ---setup=*)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---dockerfile)
+            VIASH_MODE='dockerfile'
+            shift 1
+            ;;
+        ---docker_run_args)
+            VIASH_DOCKER_RUN_ARGS+=("$2")
+            shift 2
+            ;;
+        ---docker_run_args=*)
+            VIASH_DOCKER_RUN_ARGS+=("$(ViashRemoveFlags "$1")")
+            shift 1
+            ;;
+        ---docker_image_id)
+            VIASH_MODE='docker_image_id'
+            shift 1
+            ;;
+        ---debug)
+            VIASH_MODE='debug'
+            shift 1
+            ;;
+        ---cpus)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---cpus. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---cpus=*)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus=*\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---memory)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---memory. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---memory=*)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory=*\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        *)  # positional arg or unknown option
+            # since the positional args will be eval'd, can we always quote, instead of using ViashQuote
+            VIASH_POSITIONAL_ARGS="$VIASH_POSITIONAL_ARGS '$1'"
+            [[ $1 == -* ]] && ViashWarning $1 looks like a parameter but is not a defined parameter and will instead be treated as a positional argument. Use "--help" to get more information on the parameters.
+            shift # past argument
+            ;;
+    esac
+done
+
+# parse positional parameters
+eval set -- $VIASH_POSITIONAL_ARGS
+
+
+if   [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  VIASH_ENGINE_TYPE='native'
+elif   [ "$VIASH_ENGINE_ID" == "docker" ]  ; then
+  VIASH_ENGINE_TYPE='docker'
+else
+  ViashError "Engine '$VIASH_ENGINE_ID' is not recognized. Options are: docker, native."
+  exit 1
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # check if docker is installed properly
+  ViashDockerInstallationCheck
+
+  # determine docker image id
+  if [[ "$VIASH_ENGINE_ID" == 'docker' ]]; then
+    VIASH_DOCKER_IMAGE_ID='images.viash-hub.com/vsh/openpipeline_incubator/ingestion_qc/detect_ingestion_method:build_main'
+  fi
+
+  # print dockerfile
+  if [ "$VIASH_MODE" == "dockerfile" ]; then
+    ViashDockerfile "$VIASH_ENGINE_ID"
+    exit 0
+
+  elif [ "$VIASH_MODE" == "docker_image_id" ]; then
+    echo "$VIASH_DOCKER_IMAGE_ID"
+    exit 0
+  
+  # enter docker container
+  elif [[ "$VIASH_MODE" == "debug" ]]; then
+    VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} -v '$(pwd)':/pwd --workdir /pwd -t $VIASH_DOCKER_IMAGE_ID"
+    ViashNotice "+ $VIASH_CMD"
+    eval $VIASH_CMD
+    exit 
+
+  # build docker image
+  elif [ "$VIASH_MODE" == "setup" ]; then
+    ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" "$VIASH_SETUP_STRATEGY"
+    ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+    exit 0
+  fi
+
+  # check if docker image exists
+  ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" ifneedbepullelsecachedbuild
+  ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+fi
+
+# setting computational defaults
+
+# helper function for parsing memory strings
+function ViashMemoryAsBytes {
+  local memory=`echo "$1" | tr '[:upper:]' '[:lower:]' | tr -d '[:space:]'`
+  local memory_regex='^([0-9]+)([kmgtp]i?b?|b)$'
+  if [[ $memory =~ $memory_regex ]]; then
+    local number=${memory/[^0-9]*/}
+    local symbol=${memory/*[0-9]/}
+    
+    case $symbol in
+      b)      memory_b=$number ;;
+      kb|k)   memory_b=$(( $number * 1000 )) ;;
+      mb|m)   memory_b=$(( $number * 1000 * 1000 )) ;;
+      gb|g)   memory_b=$(( $number * 1000 * 1000 * 1000 )) ;;
+      tb|t)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 )) ;;
+      pb|p)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 * 1000 )) ;;
+      kib|ki)   memory_b=$(( $number * 1024 )) ;;
+      mib|mi)   memory_b=$(( $number * 1024 * 1024 )) ;;
+      gib|gi)   memory_b=$(( $number * 1024 * 1024 * 1024 )) ;;
+      tib|ti)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 )) ;;
+      pib|pi)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 * 1024 )) ;;
+    esac
+    echo "$memory_b"
+  fi
+}
+# compute memory in different units
+if [ ! -z ${VIASH_META_MEMORY+x} ]; then
+  VIASH_META_MEMORY_B=`ViashMemoryAsBytes $VIASH_META_MEMORY`
+  # do not define other variables if memory_b is an empty string
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_META_MEMORY_KB=$(( ($VIASH_META_MEMORY_B+999) / 1000 ))
+    VIASH_META_MEMORY_MB=$(( ($VIASH_META_MEMORY_KB+999) / 1000 ))
+    VIASH_META_MEMORY_GB=$(( ($VIASH_META_MEMORY_MB+999) / 1000 ))
+    VIASH_META_MEMORY_TB=$(( ($VIASH_META_MEMORY_GB+999) / 1000 ))
+    VIASH_META_MEMORY_PB=$(( ($VIASH_META_MEMORY_TB+999) / 1000 ))
+    VIASH_META_MEMORY_KIB=$(( ($VIASH_META_MEMORY_B+1023) / 1024 ))
+    VIASH_META_MEMORY_MIB=$(( ($VIASH_META_MEMORY_KIB+1023) / 1024 ))
+    VIASH_META_MEMORY_GIB=$(( ($VIASH_META_MEMORY_MIB+1023) / 1024 ))
+    VIASH_META_MEMORY_TIB=$(( ($VIASH_META_MEMORY_GIB+1023) / 1024 ))
+    VIASH_META_MEMORY_PIB=$(( ($VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  else
+    # unset memory if string is empty
+    unset $VIASH_META_MEMORY_B
+  fi
+fi
+# unset nproc if string is empty
+if [ -z "$VIASH_META_CPUS" ]; then
+  unset $VIASH_META_CPUS
+fi
+
+
+# check whether required parameters exist
+if [ -z ${VIASH_PAR_INPUT+x} ]; then
+  ViashError '--input' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_OUTPUT+x} ]; then
+  ViashError '--output' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_NAME+x} ]; then
+  ViashError 'name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then
+  ViashError 'functionality_name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_RESOURCES_DIR+x} ]; then
+  ViashError 'resources_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_EXECUTABLE+x} ]; then
+  ViashError 'executable' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_CONFIG+x} ]; then
+  ViashError 'config' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_TEMP_DIR+x} ]; then
+  ViashError 'temp_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+
+# filling in defaults
+if [ -z ${VIASH_PAR_MODALITY+x} ]; then
+  VIASH_PAR_MODALITY="rna"
+fi
+if [ -z ${VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD+x} ]; then
+  VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD="ingestion_method"
+fi
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_INPUT" ] && [ ! -e "$VIASH_PAR_INPUT" ]; then
+  ViashError "Input file '$VIASH_PAR_INPUT' does not exist."
+  exit 1
+fi
+
+# check whether parameters values are of the right type
+if [[ -n "$VIASH_META_CPUS" ]]; then
+  if ! [[ "$VIASH_META_CPUS" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'cpus' has to be an integer. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_B" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_B" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_b' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+
+# create parent directories of output files, if so desired
+if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -d "$(dirname "$VIASH_PAR_OUTPUT")" ]; then
+  mkdir -p "$(dirname "$VIASH_PAR_OUTPUT")"
+fi
+
+if  [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  if [ "$VIASH_MODE" == "run" ]; then
+    VIASH_CMD="bash"
+  else
+    ViashError "Engine '$VIASH_ENGINE_ID' does not support mode '$VIASH_MODE'."
+    exit 1
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # detect volumes from file arguments
+  VIASH_CHOWN_VARS=()
+if [ ! -z "$VIASH_PAR_INPUT" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_INPUT")" )
+  VIASH_PAR_INPUT=$(ViashDockerAutodetectMount "$VIASH_PAR_INPUT")
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_OUTPUT")" )
+  VIASH_PAR_OUTPUT=$(ViashDockerAutodetectMount "$VIASH_PAR_OUTPUT")
+  VIASH_CHOWN_VARS+=( "$VIASH_PAR_OUTPUT" )
+fi
+if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_RESOURCES_DIR")" )
+  VIASH_META_RESOURCES_DIR=$(ViashDockerAutodetectMount "$VIASH_META_RESOURCES_DIR")
+fi
+if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_EXECUTABLE")" )
+  VIASH_META_EXECUTABLE=$(ViashDockerAutodetectMount "$VIASH_META_EXECUTABLE")
+fi
+if [ ! -z "$VIASH_META_CONFIG" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_CONFIG")" )
+  VIASH_META_CONFIG=$(ViashDockerAutodetectMount "$VIASH_META_CONFIG")
+fi
+if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_TEMP_DIR")" )
+  VIASH_META_TEMP_DIR=$(ViashDockerAutodetectMount "$VIASH_META_TEMP_DIR")
+fi
+  
+  # get unique mounts
+  VIASH_UNIQUE_MOUNTS=($(for val in "${VIASH_DIRECTORY_MOUNTS[@]}"; do echo "$val"; done | sort -u))
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # change file ownership
+  function ViashPerformChown {
+    if (( ${#VIASH_CHOWN_VARS[@]} )); then
+      set +e
+      VIASH_CMD="docker run --entrypoint=bash --rm ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID -c 'chown $(id -u):$(id -g) --silent --recursive ${VIASH_CHOWN_VARS[@]}'"
+      ViashDebug "+ $VIASH_CMD"
+      eval $VIASH_CMD
+      set -e
+    fi
+  }
+  trap ViashPerformChown EXIT
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # helper function for filling in extra docker args
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--memory=${VIASH_META_MEMORY_B}")
+  fi
+  if [ ! -z "$VIASH_META_CPUS" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--cpus=${VIASH_META_CPUS}")
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID"
+fi
+
+
+# set dependency paths
+
+
+ViashDebug "Running command: $(echo $VIASH_CMD)"
+cat << VIASHEOF | eval $VIASH_CMD
+set -e
+tempscript=\$(mktemp "$VIASH_META_TEMP_DIR/viash-run-detect_ingestion_method-XXXXXX").py
+function clean_up {
+  rm "\$tempscript"
+}
+function interrupt {
+  echo -e "\nCTRL-C Pressed..."
+  exit 1
+}
+trap clean_up EXIT
+trap interrupt INT SIGINT
+cat > "\$tempscript" << 'VIASHMAIN'
+import shutil
+import anndata as ad
+import h5py
+import sys
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'output_uns_ingestion_method': $( if [ ! -z ${VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD+x} ]; then echo "r'${VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\'/\'\"\'\"r\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\'/\'\"\'\"r\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main(par):
+
+    # read h5mu file
+    with h5py.File(par["input"], "r") as file:
+        mod = file["mod"][par["modality"]]
+        uns = ad.experimental.read_elem(file["uns"])
+        mod_obs = ad.experimental.read_elem(mod["obs"])
+        mod_uns = ad.experimental.read_elem(mod["uns"])
+
+    # detect ingestion method
+    ingestion_methods = {
+            "cellranger_multi": "metrics_cellranger" in uns,
+            "xenium": all(key in mod_obs for key in ["segmentation_method", "nucleus_area"]),
+            "cosmx": "spatial" in mod_uns and "fov" in mod_obs
+        }
+
+    # make sure only one ingestion method is detected
+    detected_methods = [method for method, detected in ingestion_methods.items() if detected]
+    methods_count = len(detected_methods)
+
+    if methods_count == 1:
+        detected_method = detected_methods[0]
+        logger.info(f"Detected ingestion method {detected_method}")
+
+    elif methods_count == 0:
+        raise ValueError("No ingestion method detected")
+
+    else:
+        raise ValueError(f"Multiple ingestion methods detected: {', '.join(detected_methods)}")
+
+    # check if mod_uns already contains a different detected method
+    if mod_uns.get(par["output_uns_ingestion_method"], detected_method) != detected_method:
+        raise ValueError(f"Field .uns['{par['output_uns_ingestion_method']}'] already exists and contains different value \`{mod_uns.get(par['output_uns_ingestion_method'])}\` than detected method (\`{detected_method}\`).")
+
+    # copy input to output
+    shutil.copy(par["input"], par["output"])
+
+    if par["output_uns_ingestion_method"] not in mod_uns:
+        with h5py.File (par["output"], "r+") as out_file:
+            out_file["uns"][par["output_uns_ingestion_method"]] = detected_method
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "\$tempscript" &
+wait "\$!"
+
+VIASHEOF
+
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # strip viash automount from file paths
+  
+  if [ ! -z "$VIASH_PAR_INPUT" ]; then
+    VIASH_PAR_INPUT=$(ViashDockerStripAutomount "$VIASH_PAR_INPUT")
+  fi
+  if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
+    VIASH_PAR_OUTPUT=$(ViashDockerStripAutomount "$VIASH_PAR_OUTPUT")
+  fi
+  if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+    VIASH_META_RESOURCES_DIR=$(ViashDockerStripAutomount "$VIASH_META_RESOURCES_DIR")
+  fi
+  if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+    VIASH_META_EXECUTABLE=$(ViashDockerStripAutomount "$VIASH_META_EXECUTABLE")
+  fi
+  if [ ! -z "$VIASH_META_CONFIG" ]; then
+    VIASH_META_CONFIG=$(ViashDockerStripAutomount "$VIASH_META_CONFIG")
+  fi
+  if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+    VIASH_META_TEMP_DIR=$(ViashDockerStripAutomount "$VIASH_META_TEMP_DIR")
+  fi
+fi
+
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -e "$VIASH_PAR_OUTPUT" ]; then
+  ViashError "Output file '$VIASH_PAR_OUTPUT' does not exist."
+  exit 1
+fi
+
+
+exit 0
diff --git a/target/executable/ingestion_qc/detect_ingestion_method/nextflow_labels.config b/target/executable/ingestion_qc/detect_ingestion_method/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/executable/ingestion_qc/detect_ingestion_method/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/executable/ingestion_qc/detect_ingestion_method/setup_logger.py b/target/executable/ingestion_qc/detect_ingestion_method/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/executable/ingestion_qc/detect_ingestion_method/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/executable/ingestion_qc/generate_html/.config.vsh.yaml b/target/executable/ingestion_qc/generate_html/.config.vsh.yaml
new file mode 100644
index 0000000..547a06c
--- /dev/null
+++ b/target/executable/ingestion_qc/generate_html/.config.vsh.yaml
@@ -0,0 +1,267 @@
+name: "generate_html"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input_data"
+    description: "The input JSON file containing the QC metrics"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input_structure"
+    description: "The input JSON file containing the structure of the data"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_qc_report"
+    description: "The output HTML report"
+    info: null
+    example:
+    - "path/to/file.html"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Generate an HTML report from the QC metrics"
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+- type: "file"
+  path: "sc_dataset.json"
+- type: "file"
+  path: "sc_report_structure.json"
+- type: "file"
+  path: "xenium_dataset.json"
+- type: "file"
+  path: "xenium_report_structure.json"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "node:latest"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "docker"
+    run:
+    - "npm install -g pnpm@latest-10 \\\n&& cd /opt && git clone https://github.com/openpipelines-bio/incubator_ingestion_qc.git\
+      \ \\\n&& cd incubator_ingestion_qc && pnpm install \\\n&& true\n"
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/generate_html/config.vsh.yaml"
+  runner: "executable"
+  engine: "docker|native"
+  output: "target/executable/ingestion_qc/generate_html"
+  executable: "target/executable/ingestion_qc/generate_html/generate_html"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/executable/ingestion_qc/generate_html/generate_html b/target/executable/ingestion_qc/generate_html/generate_html
new file mode 100755
index 0000000..ab5d4ff
--- /dev/null
+++ b/target/executable/ingestion_qc/generate_html/generate_html
@@ -0,0 +1,1163 @@
+#!/usr/bin/env bash
+
+# generate_html build_main
+# 
+# This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+# work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+# Intuitive.
+# 
+# The component may contain files which fall under a different license. The
+# authors of this component should specify the license in the header of such
+# files, or include a separate license file detailing the licenses of all included
+# files.
+# 
+# Component authors:
+#  * Jakub Majercik (author)
+#  * Dorien Roosen (author)
+#  * Robrecht Cannoodt (author)
+#  * Weiwei Schultz (contributor)
+
+set -e
+
+if [ -z "$VIASH_TEMP" ]; then
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TMP}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMPDIR}
+  VIASH_TEMP=${VIASH_TEMP:-$TEMP}
+  VIASH_TEMP=${VIASH_TEMP:-/tmp}
+fi
+
+# define helper functions
+# ViashQuote: put quotes around non flag values
+# $1     : unquoted string
+# return : possibly quoted string
+# examples:
+#   ViashQuote --foo      # returns --foo
+#   ViashQuote bar        # returns 'bar'
+#   Viashquote --foo=bar  # returns --foo='bar'
+function ViashQuote {
+  if [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+=.+$ ]]; then
+    echo "$1" | sed "s#=\(.*\)#='\1'#"
+  elif [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+$ ]]; then
+    echo "$1"
+  else
+    echo "'$1'"
+  fi
+}
+# ViashRemoveFlags: Remove leading flag
+# $1     : string with a possible leading flag
+# return : string without possible leading flag
+# examples:
+#   ViashRemoveFlags --foo=bar  # returns bar
+function ViashRemoveFlags {
+  echo "$1" | sed 's/^--*[a-zA-Z0-9_\-]*=//'
+}
+# ViashSourceDir: return the path of a bash file, following symlinks
+# usage   : ViashSourceDir ${BASH_SOURCE[0]}
+# $1      : Should always be set to ${BASH_SOURCE[0]}
+# returns : The absolute path of the bash file
+function ViashSourceDir {
+  local source="$1"
+  while [ -h "$source" ]; do
+    local dir="$( cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd )"
+    source="$(readlink "$source")"
+    [[ $source != /* ]] && source="$dir/$source"
+  done
+  cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd
+}
+# ViashFindTargetDir: return the path of the '.build.yaml' file, following symlinks
+# usage   : ViashFindTargetDir 'ScriptPath'
+# $1      : The location from where to start the upward search
+# returns : The absolute path of the '.build.yaml' file
+function ViashFindTargetDir {
+  local source="$1"
+  while [[ "$source" != "" && ! -e "$source/.build.yaml" ]]; do
+    source=${source%/*}
+  done
+  echo $source
+}
+# see https://en.wikipedia.org/wiki/Syslog#Severity_level
+VIASH_LOGCODE_EMERGENCY=0
+VIASH_LOGCODE_ALERT=1
+VIASH_LOGCODE_CRITICAL=2
+VIASH_LOGCODE_ERROR=3
+VIASH_LOGCODE_WARNING=4
+VIASH_LOGCODE_NOTICE=5
+VIASH_LOGCODE_INFO=6
+VIASH_LOGCODE_DEBUG=7
+VIASH_VERBOSITY=$VIASH_LOGCODE_NOTICE
+
+# ViashLog: Log events depending on the verbosity level
+# usage: ViashLog 1 alert Oh no something went wrong!
+# $1: required verbosity level
+# $2: display tag
+# $3+: messages to display
+# stdout: Your input, prepended by '[$2] '.
+function ViashLog {
+  local required_level="$1"
+  local display_tag="$2"
+  shift 2
+  if [ $VIASH_VERBOSITY -ge $required_level ]; then
+    >&2 echo "[$display_tag]" "$@"
+  fi
+}
+
+# ViashEmergency: log events when the system is unstable
+# usage: ViashEmergency Oh no something went wrong.
+# stdout: Your input, prepended by '[emergency] '.
+function ViashEmergency {
+  ViashLog $VIASH_LOGCODE_EMERGENCY emergency "$@"
+}
+
+# ViashAlert: log events when actions must be taken immediately (e.g. corrupted system database)
+# usage: ViashAlert Oh no something went wrong.
+# stdout: Your input, prepended by '[alert] '.
+function ViashAlert {
+  ViashLog $VIASH_LOGCODE_ALERT alert "$@"
+}
+
+# ViashCritical: log events when a critical condition occurs
+# usage: ViashCritical Oh no something went wrong.
+# stdout: Your input, prepended by '[critical] '.
+function ViashCritical {
+  ViashLog $VIASH_LOGCODE_CRITICAL critical "$@"
+}
+
+# ViashError: log events when an error condition occurs
+# usage: ViashError Oh no something went wrong.
+# stdout: Your input, prepended by '[error] '.
+function ViashError {
+  ViashLog $VIASH_LOGCODE_ERROR error "$@"
+}
+
+# ViashWarning: log potentially abnormal events
+# usage: ViashWarning Something may have gone wrong.
+# stdout: Your input, prepended by '[warning] '.
+function ViashWarning {
+  ViashLog $VIASH_LOGCODE_WARNING warning "$@"
+}
+
+# ViashNotice: log significant but normal events
+# usage: ViashNotice This just happened.
+# stdout: Your input, prepended by '[notice] '.
+function ViashNotice {
+  ViashLog $VIASH_LOGCODE_NOTICE notice "$@"
+}
+
+# ViashInfo: log normal events
+# usage: ViashInfo This just happened.
+# stdout: Your input, prepended by '[info] '.
+function ViashInfo {
+  ViashLog $VIASH_LOGCODE_INFO info "$@"
+}
+
+# ViashDebug: log all events, for debugging purposes
+# usage: ViashDebug This just happened.
+# stdout: Your input, prepended by '[debug] '.
+function ViashDebug {
+  ViashLog $VIASH_LOGCODE_DEBUG debug "$@"
+}
+
+# find source folder of this component
+VIASH_META_RESOURCES_DIR=`ViashSourceDir ${BASH_SOURCE[0]}`
+
+# find the root of the built components & dependencies
+VIASH_TARGET_DIR=`ViashFindTargetDir $VIASH_META_RESOURCES_DIR`
+
+# define meta fields
+VIASH_META_NAME="generate_html"
+VIASH_META_FUNCTIONALITY_NAME="generate_html"
+VIASH_META_EXECUTABLE="$VIASH_META_RESOURCES_DIR/$VIASH_META_NAME"
+VIASH_META_CONFIG="$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+VIASH_META_TEMP_DIR="$VIASH_TEMP"
+
+
+
+# initialise variables
+VIASH_MODE='run'
+VIASH_ENGINE_ID='docker'
+
+######## Helper functions for setting up Docker images for viash ########
+# expects: ViashDockerBuild
+
+# ViashDockerInstallationCheck: check whether Docker is installed correctly
+#
+# examples:
+#   ViashDockerInstallationCheck
+function ViashDockerInstallationCheck {
+  ViashDebug "Checking whether Docker is installed"
+  if [ ! command -v docker &> /dev/null ]; then
+    ViashCritical "Docker doesn't seem to be installed. See 'https://docs.docker.com/get-docker/' for instructions."
+    exit 1
+  fi
+
+  ViashDebug "Checking whether the Docker daemon is running"
+  local save=$-; set +e
+  local docker_version=$(docker version --format '{{.Client.APIVersion}}' 2> /dev/null)
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashCritical "Docker daemon does not seem to be running. Try one of the following:"
+    ViashCritical "- Try running 'dockerd' in the command line"
+    ViashCritical "- See https://docs.docker.com/config/daemon/"
+    exit 1
+  fi
+}
+
+# ViashDockerRemoteTagCheck: check whether a Docker image is available 
+# on a remote. Assumes `docker login` has been performed, if relevant.
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerRemoteTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerRemoteTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerRemoteTagCheck {
+  docker manifest inspect $1 > /dev/null 2> /dev/null
+}
+
+# ViashDockerLocalTagCheck: check whether a Docker image is available locally
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   docker pull python:latest
+#   ViashDockerLocalTagCheck python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerLocalTagCheck sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerLocalTagCheck {
+  [ -n "$(docker images -q $1)" ]
+}
+
+# ViashDockerPull: pull a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPull python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPull sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPull {
+  ViashNotice "Checking if Docker image is available at '$1'"
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker pull $1 && return 0 || return 1
+  else
+    local save=$-; set +e
+    docker pull $1 2> /dev/null > /dev/null
+    local out=$?
+    [[ $save =~ e ]] && set -e
+    if [ $out -ne 0 ]; then
+      ViashWarning "Could not pull from '$1'. Docker image doesn't exist or is not accessible."
+    fi
+    return $out
+  fi
+}
+
+# ViashDockerPush: push a Docker image
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# exit code $?        : whether or not the image was found
+# examples:
+#   ViashDockerPush python:latest
+#   echo $?                                     # returns '0'
+#   ViashDockerPush sdaizudceahifu
+#   echo $?                                     # returns '1'
+function ViashDockerPush {
+  ViashNotice "Pushing image to '$1'"
+  local save=$-; set +e
+  local out
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    docker push $1
+    out=$?
+  else
+    docker push $1 2> /dev/null > /dev/null
+    out=$?
+  fi
+  [[ $save =~ e ]] && set -e
+  if [ $out -eq 0 ]; then
+    ViashNotice "Container '$1' push succeeded."
+  else
+    ViashError "Container '$1' push errored. You might not be logged in or have the necessary permissions."
+  fi
+  return $out
+}
+
+# ViashDockerPullElseBuild: pull a Docker image, else build it
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# ViashDockerBuild    : a Bash function which builds a docker image, takes image identifier as argument.
+# examples:
+#   ViashDockerPullElseBuild mynewcomponent
+function ViashDockerPullElseBuild {
+  local save=$-; set +e
+  ViashDockerPull $1
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashDockerBuild $@
+  fi
+}
+
+# ViashDockerSetup: create a Docker image, according to specified docker setup strategy
+#
+# $1          : image identifier with format `[registry/]image[:tag]`
+# $2          : docker setup strategy, see DockerSetupStrategy.scala
+# examples:
+#   ViashDockerSetup mynewcomponent alwaysbuild
+function ViashDockerSetup {
+  local image_id="$1"
+  local setup_strategy="$2"
+  if [ "$setup_strategy" == "alwaysbuild" -o "$setup_strategy" == "build" -o "$setup_strategy" == "b" ]; then
+    ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspull" -o "$setup_strategy" == "pull" -o "$setup_strategy" == "p" ]; then
+    ViashDockerPull $image_id
+  elif [ "$setup_strategy" == "alwayspullelsebuild" -o "$setup_strategy" == "pullelsebuild" ]; then
+    ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayspullelsecachedbuild" -o "$setup_strategy" == "pullelsecachedbuild" ]; then
+    ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [ "$setup_strategy" == "alwayscachedbuild" -o "$setup_strategy" == "cachedbuild" -o "$setup_strategy" == "cb" ]; then
+    ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+  elif [[ "$setup_strategy" =~ ^ifneedbe ]]; then
+    local save=$-; set +e
+    ViashDockerLocalTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashInfo "Image $image_id already exists"
+    elif [ "$setup_strategy" == "ifneedbebuild" ]; then
+      ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbecachedbuild" ]; then
+      ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepull" ]; then
+      ViashDockerPull $image_id
+    elif [ "$setup_strategy" == "ifneedbepullelsebuild" ]; then
+      ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
+    elif [ "$setup_strategy" == "ifneedbepullelsecachedbuild" ]; then
+      ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
+    else
+      ViashError "Unrecognised Docker strategy: $setup_strategy"
+      exit 1
+    fi
+  elif [ "$setup_strategy" == "push" -o "$setup_strategy" == "forcepush" -o "$setup_strategy" == "alwayspush" ]; then
+    ViashDockerPush "$image_id"
+  elif [ "$setup_strategy" == "pushifnotpresent" -o "$setup_strategy" == "gentlepush" -o "$setup_strategy" == "maybepush" ]; then
+    local save=$-; set +e
+    ViashDockerRemoteTagCheck $image_id
+    local outCheck=$?
+    [[ $save =~ e ]] && set -e
+    if [ $outCheck -eq 0 ]; then
+      ViashNotice "Container '$image_id' exists, doing nothing."
+    else
+      ViashNotice "Container '$image_id' does not yet exist."
+      ViashDockerPush "$image_id"
+    fi
+  elif [ "$setup_strategy" == "donothing" -o "$setup_strategy" == "meh" ]; then
+    ViashNotice "Skipping setup."
+  else
+    ViashError "Unrecognised Docker strategy: $setup_strategy"
+    exit 1
+  fi
+}
+
+# ViashDockerCheckCommands: Check whether a docker container has the required commands
+#
+# $1                  : image identifier with format `[registry/]image[:tag]`
+# $@                  : commands to verify being present
+# examples:
+#   ViashDockerCheckCommands bash:4.0 bash ps foo
+function ViashDockerCheckCommands {
+  local image_id="$1"
+  shift 1
+  local commands="$@"
+  local save=$-; set +e
+  local missing # mark 'missing' as local in advance, otherwise the exit code of the command will be missing and always be '0'
+  missing=$(docker run --rm --entrypoint=sh "$image_id" -c "for command in $commands; do command -v \$command >/dev/null 2>&1; if [ \$? -ne 0 ]; then echo \$command; exit 1; fi; done")
+  local outCheck=$?
+  [[ $save =~ e ]] && set -e
+  if [ $outCheck -ne 0 ]; then
+  	ViashError "Docker container '$image_id' does not contain command '$missing'."
+  	exit 1
+  fi
+}
+
+# ViashDockerBuild: build a docker image
+# $1                               : image identifier with format `[registry/]image[:tag]`
+# $...                             : additional arguments to pass to docker build
+# $VIASH_META_TEMP_DIR             : temporary directory to store dockerfile & optional resources in
+# $VIASH_META_NAME                 : name of the component
+# $VIASH_META_RESOURCES_DIR        : directory containing the resources
+# $VIASH_VERBOSITY                 : verbosity level
+# exit code $?                     : whether or not the image was built successfully
+function ViashDockerBuild {
+  local image_id="$1"
+  shift 1
+
+  # create temporary directory to store dockerfile & optional resources in
+  local tmpdir=$(mktemp -d "$VIASH_META_TEMP_DIR/dockerbuild-$VIASH_META_NAME-XXXXXX")
+  local dockerfile="$tmpdir/Dockerfile"
+  function clean_up {
+    rm -rf "$tmpdir"
+  }
+  trap clean_up EXIT
+
+  # store dockerfile and resources
+  ViashDockerfile "$VIASH_ENGINE_ID" > "$dockerfile"
+
+  # generate the build command
+  local docker_build_cmd="docker build -t '$image_id' $@ '$VIASH_META_RESOURCES_DIR' -f '$dockerfile'"
+
+  # build the container
+  ViashNotice "Building container '$image_id' with Dockerfile"
+  ViashInfo "$docker_build_cmd"
+  local save=$-; set +e
+  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
+    eval $docker_build_cmd
+  else
+    eval $docker_build_cmd &> "$tmpdir/docker_build.log"
+  fi
+
+  # check exit code
+  local out=$?
+  [[ $save =~ e ]] && set -e
+  if [ $out -ne 0 ]; then
+    ViashError "Error occurred while building container '$image_id'"
+    if [ $VIASH_VERBOSITY -lt $VIASH_LOGCODE_INFO ]; then
+      ViashError "Transcript: --------------------------------"
+      cat "$tmpdir/docker_build.log"
+      ViashError "End of transcript --------------------------"
+    fi
+    exit 1
+  fi
+}
+
+######## End of helper functions for setting up Docker images for viash ########
+
+# ViashDockerFile: print the dockerfile to stdout
+# $1    : engine identifier
+# return : dockerfile required to run this component
+# examples:
+#   ViashDockerFile
+function ViashDockerfile {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    cat << 'VIASHDOCKER'
+FROM node:latest
+ENTRYPOINT []
+RUN apt-get update && \
+  DEBIAN_FRONTEND=noninteractive apt-get install -y git && \
+  rm -rf /var/lib/apt/lists/*
+
+RUN npm install -g pnpm@latest-10 \
+&& cd /opt && git clone https://github.com/openpipelines-bio/incubator_ingestion_qc.git \
+&& cd incubator_ingestion_qc && pnpm install \
+&& true
+
+LABEL org.opencontainers.image.authors="Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz"
+LABEL org.opencontainers.image.description="Companion container for running component ingestion_qc generate_html"
+LABEL org.opencontainers.image.created="2025-08-11T08:32:06Z"
+LABEL org.opencontainers.image.source="https://github.com/openpipelines-bio/openpipeline_incubator"
+LABEL org.opencontainers.image.revision="dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+LABEL org.opencontainers.image.version="build_main"
+
+VIASHDOCKER
+  fi
+}
+
+# ViashDockerBuildArgs: return the arguments to pass to docker build
+# $1    : engine identifier
+# return : arguments to pass to docker build
+function ViashDockerBuildArgs {
+  local engine_id="$1"
+
+  if [[ "$engine_id" == "docker" ]]; then
+    echo ""
+  fi
+}
+
+# ViashAbsolutePath: generate absolute path from relative path
+# borrowed from https://stackoverflow.com/a/21951256
+# $1     : relative filename
+# return : absolute path
+# examples:
+#   ViashAbsolutePath some_file.txt   # returns /path/to/some_file.txt
+#   ViashAbsolutePath /foo/bar/..     # returns /foo
+function ViashAbsolutePath {
+  local thePath
+  local parr
+  local outp
+  local len
+  if [[ ! "$1" =~ ^/ ]]; then
+    thePath="$PWD/$1"
+  else
+    thePath="$1"
+  fi
+  echo "$thePath" | (
+    IFS=/
+    read -a parr
+    declare -a outp
+    for i in "${parr[@]}"; do
+      case "$i" in
+      ''|.) continue ;;
+      ..)
+        len=${#outp[@]}
+        if ((len==0)); then
+          continue
+        else
+          unset outp[$((len-1))]
+        fi
+        ;;
+      *)
+        len=${#outp[@]}
+        outp[$len]="$i"
+      ;;
+      esac
+    done
+    echo /"${outp[*]}"
+  )
+}
+# ViashDockerAutodetectMount: auto configuring docker mounts from parameters
+# $1                             : The parameter value
+# returns                        : New parameter
+# $VIASH_DIRECTORY_MOUNTS        : Added another parameter to be passed to docker
+# $VIASH_DOCKER_AUTOMOUNT_PREFIX : The prefix to be used for the automounts
+# examples:
+#   ViashDockerAutodetectMount /path/to/bar      # returns '/viash_automount/path/to/bar'
+#   ViashDockerAutodetectMountArg /path/to/bar   # returns '--volume="/path/to:/viash_automount/path/to"'
+function ViashDockerAutodetectMount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  if [ -z "$base_name" ]; then
+    echo "$mount_target"
+  else
+    echo "$mount_target/$base_name"
+  fi
+}
+function ViashDockerAutodetectMountArg {
+  local abs_path=$(ViashAbsolutePath "$1")
+  local mount_source
+  local base_name
+  if [ -d "$abs_path" ]; then
+    mount_source="$abs_path"
+    base_name=""
+  else
+    mount_source=`dirname "$abs_path"`
+    base_name=`basename "$abs_path"`
+  fi
+  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
+  ViashDebug "ViashDockerAutodetectMountArg $1 -> $mount_source -> $mount_target"
+  echo "--volume=\"$mount_source:$mount_target\""
+}
+function ViashDockerStripAutomount {
+  local abs_path=$(ViashAbsolutePath "$1")
+  echo "${abs_path#$VIASH_DOCKER_AUTOMOUNT_PREFIX}"
+}
+# initialise variables
+VIASH_DIRECTORY_MOUNTS=()
+
+# configure default docker automount prefix if it is unset
+if [ -z "${VIASH_DOCKER_AUTOMOUNT_PREFIX+x}" ]; then
+  VIASH_DOCKER_AUTOMOUNT_PREFIX="/viash_automount"
+fi
+
+# initialise docker variables
+VIASH_DOCKER_RUN_ARGS=(-i --rm)
+
+
+# ViashHelp: Display helpful explanation about this executable
+function ViashHelp {
+  echo "generate_html build_main"
+  echo ""
+  echo "Generate an HTML report from the QC metrics"
+  echo ""
+  echo "Inputs:"
+  echo "    --input_data"
+  echo "        type: file, required parameter, file must exist"
+  echo "        example: path/to/file.json"
+  echo "        The input JSON file containing the QC metrics"
+  echo ""
+  echo "    --input_structure"
+  echo "        type: file, required parameter, file must exist"
+  echo "        example: path/to/file.json"
+  echo "        The input JSON file containing the structure of the data"
+  echo ""
+  echo "Outputs:"
+  echo "    --output_qc_report"
+  echo "        type: file, required parameter, output, file must exist"
+  echo "        example: path/to/file.html"
+  echo "        The output HTML report"
+  echo ""
+  echo "Viash built in Computational Requirements:"
+  echo "    ---cpus=INT"
+  echo "        Number of CPUs to use"
+  echo "    ---memory=STRING"
+  echo "        Amount of memory to use. Examples: 4GB, 3MiB."
+  echo ""
+  echo "Viash built in Docker:"
+  echo "    ---setup=STRATEGY"
+  echo "        Setup the docker container. Options are: alwaysbuild, alwayscachedbuild, ifneedbebuild, ifneedbecachedbuild, alwayspull, alwayspullelsebuild, alwayspullelsecachedbuild, ifneedbepull, ifneedbepullelsebuild, ifneedbepullelsecachedbuild, push, pushifnotpresent, donothing."
+  echo "        Default: ifneedbepullelsecachedbuild"
+  echo "    ---dockerfile"
+  echo "        Print the dockerfile to stdout."
+  echo "    ---docker_run_args=ARG"
+  echo "        Provide runtime arguments to Docker. See the documentation on \`docker run\` for more information."
+  echo "    ---docker_image_id"
+  echo "        Print the docker image id to stdout."
+  echo "    ---debug"
+  echo "        Enter the docker container for debugging purposes."
+  echo ""
+  echo "Viash built in Engines:"
+  echo "    ---engine=ENGINE_ID"
+  echo "        Specify the engine to use. Options are: docker, native."
+  echo "        Default: docker"
+}
+
+# initialise array
+VIASH_POSITIONAL_ARGS=''
+
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        -h|--help)
+            ViashHelp
+            exit
+            ;;
+        ---v|---verbose)
+            let "VIASH_VERBOSITY=VIASH_VERBOSITY+1"
+            shift 1
+            ;;
+        ---verbosity)
+            VIASH_VERBOSITY="$2"
+            shift 2
+            ;;
+        ---verbosity=*)
+            VIASH_VERBOSITY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        --version)
+            echo "generate_html build_main"
+            exit
+            ;;
+        --input_data)
+            [ -n "$VIASH_PAR_INPUT_DATA" ] && ViashError Bad arguments for option \'--input_data\': \'$VIASH_PAR_INPUT_DATA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT_DATA="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --input_data. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --input_data=*)
+            [ -n "$VIASH_PAR_INPUT_DATA" ] && ViashError Bad arguments for option \'--input_data=*\': \'$VIASH_PAR_INPUT_DATA\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT_DATA=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --input_structure)
+            [ -n "$VIASH_PAR_INPUT_STRUCTURE" ] && ViashError Bad arguments for option \'--input_structure\': \'$VIASH_PAR_INPUT_STRUCTURE\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT_STRUCTURE="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --input_structure. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --input_structure=*)
+            [ -n "$VIASH_PAR_INPUT_STRUCTURE" ] && ViashError Bad arguments for option \'--input_structure=*\': \'$VIASH_PAR_INPUT_STRUCTURE\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_INPUT_STRUCTURE=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        --output_qc_report)
+            [ -n "$VIASH_PAR_OUTPUT_QC_REPORT" ] && ViashError Bad arguments for option \'--output_qc_report\': \'$VIASH_PAR_OUTPUT_QC_REPORT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_QC_REPORT="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_qc_report. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        --output_qc_report=*)
+            [ -n "$VIASH_PAR_OUTPUT_QC_REPORT" ] && ViashError Bad arguments for option \'--output_qc_report=*\': \'$VIASH_PAR_OUTPUT_QC_REPORT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_PAR_OUTPUT_QC_REPORT=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---engine)
+            VIASH_ENGINE_ID="$2"
+            shift 2
+            ;;
+        ---engine=*)
+            VIASH_ENGINE_ID="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---setup)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$2"
+            shift 2
+            ;;
+        ---setup=*)
+            VIASH_MODE='setup'
+            VIASH_SETUP_STRATEGY="$(ViashRemoveFlags "$1")"
+            shift 1
+            ;;
+        ---dockerfile)
+            VIASH_MODE='dockerfile'
+            shift 1
+            ;;
+        ---docker_run_args)
+            VIASH_DOCKER_RUN_ARGS+=("$2")
+            shift 2
+            ;;
+        ---docker_run_args=*)
+            VIASH_DOCKER_RUN_ARGS+=("$(ViashRemoveFlags "$1")")
+            shift 1
+            ;;
+        ---docker_image_id)
+            VIASH_MODE='docker_image_id'
+            shift 1
+            ;;
+        ---debug)
+            VIASH_MODE='debug'
+            shift 1
+            ;;
+        ---cpus)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---cpus. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---cpus=*)
+            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus=*\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_CPUS=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        ---memory)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY="$2"
+            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---memory. Use "--help" to get more information on the parameters. && exit 1
+            shift 2
+            ;;
+        ---memory=*)
+            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory=*\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
+            VIASH_META_MEMORY=$(ViashRemoveFlags "$1")
+            shift 1
+            ;;
+        *)  # positional arg or unknown option
+            # since the positional args will be eval'd, can we always quote, instead of using ViashQuote
+            VIASH_POSITIONAL_ARGS="$VIASH_POSITIONAL_ARGS '$1'"
+            [[ $1 == -* ]] && ViashWarning $1 looks like a parameter but is not a defined parameter and will instead be treated as a positional argument. Use "--help" to get more information on the parameters.
+            shift # past argument
+            ;;
+    esac
+done
+
+# parse positional parameters
+eval set -- $VIASH_POSITIONAL_ARGS
+
+
+if   [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  VIASH_ENGINE_TYPE='native'
+elif   [ "$VIASH_ENGINE_ID" == "docker" ]  ; then
+  VIASH_ENGINE_TYPE='docker'
+else
+  ViashError "Engine '$VIASH_ENGINE_ID' is not recognized. Options are: docker, native."
+  exit 1
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # check if docker is installed properly
+  ViashDockerInstallationCheck
+
+  # determine docker image id
+  if [[ "$VIASH_ENGINE_ID" == 'docker' ]]; then
+    VIASH_DOCKER_IMAGE_ID='images.viash-hub.com/vsh/openpipeline_incubator/ingestion_qc/generate_html:build_main'
+  fi
+
+  # print dockerfile
+  if [ "$VIASH_MODE" == "dockerfile" ]; then
+    ViashDockerfile "$VIASH_ENGINE_ID"
+    exit 0
+
+  elif [ "$VIASH_MODE" == "docker_image_id" ]; then
+    echo "$VIASH_DOCKER_IMAGE_ID"
+    exit 0
+  
+  # enter docker container
+  elif [[ "$VIASH_MODE" == "debug" ]]; then
+    VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} -v '$(pwd)':/pwd --workdir /pwd -t $VIASH_DOCKER_IMAGE_ID"
+    ViashNotice "+ $VIASH_CMD"
+    eval $VIASH_CMD
+    exit 
+
+  # build docker image
+  elif [ "$VIASH_MODE" == "setup" ]; then
+    ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" "$VIASH_SETUP_STRATEGY"
+    ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+    exit 0
+  fi
+
+  # check if docker image exists
+  ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" ifneedbepullelsecachedbuild
+  ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'ps' 'bash'
+fi
+
+# setting computational defaults
+
+# helper function for parsing memory strings
+function ViashMemoryAsBytes {
+  local memory=`echo "$1" | tr '[:upper:]' '[:lower:]' | tr -d '[:space:]'`
+  local memory_regex='^([0-9]+)([kmgtp]i?b?|b)$'
+  if [[ $memory =~ $memory_regex ]]; then
+    local number=${memory/[^0-9]*/}
+    local symbol=${memory/*[0-9]/}
+    
+    case $symbol in
+      b)      memory_b=$number ;;
+      kb|k)   memory_b=$(( $number * 1000 )) ;;
+      mb|m)   memory_b=$(( $number * 1000 * 1000 )) ;;
+      gb|g)   memory_b=$(( $number * 1000 * 1000 * 1000 )) ;;
+      tb|t)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 )) ;;
+      pb|p)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 * 1000 )) ;;
+      kib|ki)   memory_b=$(( $number * 1024 )) ;;
+      mib|mi)   memory_b=$(( $number * 1024 * 1024 )) ;;
+      gib|gi)   memory_b=$(( $number * 1024 * 1024 * 1024 )) ;;
+      tib|ti)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 )) ;;
+      pib|pi)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 * 1024 )) ;;
+    esac
+    echo "$memory_b"
+  fi
+}
+# compute memory in different units
+if [ ! -z ${VIASH_META_MEMORY+x} ]; then
+  VIASH_META_MEMORY_B=`ViashMemoryAsBytes $VIASH_META_MEMORY`
+  # do not define other variables if memory_b is an empty string
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_META_MEMORY_KB=$(( ($VIASH_META_MEMORY_B+999) / 1000 ))
+    VIASH_META_MEMORY_MB=$(( ($VIASH_META_MEMORY_KB+999) / 1000 ))
+    VIASH_META_MEMORY_GB=$(( ($VIASH_META_MEMORY_MB+999) / 1000 ))
+    VIASH_META_MEMORY_TB=$(( ($VIASH_META_MEMORY_GB+999) / 1000 ))
+    VIASH_META_MEMORY_PB=$(( ($VIASH_META_MEMORY_TB+999) / 1000 ))
+    VIASH_META_MEMORY_KIB=$(( ($VIASH_META_MEMORY_B+1023) / 1024 ))
+    VIASH_META_MEMORY_MIB=$(( ($VIASH_META_MEMORY_KIB+1023) / 1024 ))
+    VIASH_META_MEMORY_GIB=$(( ($VIASH_META_MEMORY_MIB+1023) / 1024 ))
+    VIASH_META_MEMORY_TIB=$(( ($VIASH_META_MEMORY_GIB+1023) / 1024 ))
+    VIASH_META_MEMORY_PIB=$(( ($VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  else
+    # unset memory if string is empty
+    unset $VIASH_META_MEMORY_B
+  fi
+fi
+# unset nproc if string is empty
+if [ -z "$VIASH_META_CPUS" ]; then
+  unset $VIASH_META_CPUS
+fi
+
+
+# check whether required parameters exist
+if [ -z ${VIASH_PAR_INPUT_DATA+x} ]; then
+  ViashError '--input_data' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_INPUT_STRUCTURE+x} ]; then
+  ViashError '--input_structure' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_PAR_OUTPUT_QC_REPORT+x} ]; then
+  ViashError '--output_qc_report' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_NAME+x} ]; then
+  ViashError 'name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then
+  ViashError 'functionality_name' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_RESOURCES_DIR+x} ]; then
+  ViashError 'resources_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_EXECUTABLE+x} ]; then
+  ViashError 'executable' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_CONFIG+x} ]; then
+  ViashError 'config' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+if [ -z ${VIASH_META_TEMP_DIR+x} ]; then
+  ViashError 'temp_dir' is a required argument. Use "--help" to get more information on the parameters.
+  exit 1
+fi
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_INPUT_DATA" ] && [ ! -e "$VIASH_PAR_INPUT_DATA" ]; then
+  ViashError "Input file '$VIASH_PAR_INPUT_DATA' does not exist."
+  exit 1
+fi
+if [ ! -z "$VIASH_PAR_INPUT_STRUCTURE" ] && [ ! -e "$VIASH_PAR_INPUT_STRUCTURE" ]; then
+  ViashError "Input file '$VIASH_PAR_INPUT_STRUCTURE' does not exist."
+  exit 1
+fi
+
+# check whether parameters values are of the right type
+if [[ -n "$VIASH_META_CPUS" ]]; then
+  if ! [[ "$VIASH_META_CPUS" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'cpus' has to be an integer. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_B" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_B" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_b' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pb' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_KIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_KIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_kib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_MIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_MIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_mib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_GIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_GIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_gib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_TIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_TIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_tib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+if [[ -n "$VIASH_META_MEMORY_PIB" ]]; then
+  if ! [[ "$VIASH_META_MEMORY_PIB" =~ ^[-+]?[0-9]+$ ]]; then
+    ViashError 'memory_pib' has to be a long. Use "--help" to get more information on the parameters.
+    exit 1
+  fi
+fi
+
+# create parent directories of output files, if so desired
+if [ ! -z "$VIASH_PAR_OUTPUT_QC_REPORT" ] && [ ! -d "$(dirname "$VIASH_PAR_OUTPUT_QC_REPORT")" ]; then
+  mkdir -p "$(dirname "$VIASH_PAR_OUTPUT_QC_REPORT")"
+fi
+
+if  [ "$VIASH_ENGINE_ID" == "native" ]  ; then
+  if [ "$VIASH_MODE" == "run" ]; then
+    VIASH_CMD="bash"
+  else
+    ViashError "Engine '$VIASH_ENGINE_ID' does not support mode '$VIASH_MODE'."
+    exit 1
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # detect volumes from file arguments
+  VIASH_CHOWN_VARS=()
+if [ ! -z "$VIASH_PAR_INPUT_DATA" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_INPUT_DATA")" )
+  VIASH_PAR_INPUT_DATA=$(ViashDockerAutodetectMount "$VIASH_PAR_INPUT_DATA")
+fi
+if [ ! -z "$VIASH_PAR_INPUT_STRUCTURE" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_INPUT_STRUCTURE")" )
+  VIASH_PAR_INPUT_STRUCTURE=$(ViashDockerAutodetectMount "$VIASH_PAR_INPUT_STRUCTURE")
+fi
+if [ ! -z "$VIASH_PAR_OUTPUT_QC_REPORT" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_OUTPUT_QC_REPORT")" )
+  VIASH_PAR_OUTPUT_QC_REPORT=$(ViashDockerAutodetectMount "$VIASH_PAR_OUTPUT_QC_REPORT")
+  VIASH_CHOWN_VARS+=( "$VIASH_PAR_OUTPUT_QC_REPORT" )
+fi
+if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_RESOURCES_DIR")" )
+  VIASH_META_RESOURCES_DIR=$(ViashDockerAutodetectMount "$VIASH_META_RESOURCES_DIR")
+fi
+if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_EXECUTABLE")" )
+  VIASH_META_EXECUTABLE=$(ViashDockerAutodetectMount "$VIASH_META_EXECUTABLE")
+fi
+if [ ! -z "$VIASH_META_CONFIG" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_CONFIG")" )
+  VIASH_META_CONFIG=$(ViashDockerAutodetectMount "$VIASH_META_CONFIG")
+fi
+if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_TEMP_DIR")" )
+  VIASH_META_TEMP_DIR=$(ViashDockerAutodetectMount "$VIASH_META_TEMP_DIR")
+fi
+  
+  # get unique mounts
+  VIASH_UNIQUE_MOUNTS=($(for val in "${VIASH_DIRECTORY_MOUNTS[@]}"; do echo "$val"; done | sort -u))
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # change file ownership
+  function ViashPerformChown {
+    if (( ${#VIASH_CHOWN_VARS[@]} )); then
+      set +e
+      VIASH_CMD="docker run --entrypoint=bash --rm ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID -c 'chown $(id -u):$(id -g) --silent --recursive ${VIASH_CHOWN_VARS[@]}'"
+      ViashDebug "+ $VIASH_CMD"
+      eval $VIASH_CMD
+      set -e
+    fi
+  }
+  trap ViashPerformChown EXIT
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # helper function for filling in extra docker args
+  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--memory=${VIASH_META_MEMORY_B}")
+  fi
+  if [ ! -z "$VIASH_META_CPUS" ]; then
+    VIASH_DOCKER_RUN_ARGS+=("--cpus=${VIASH_META_CPUS}")
+  fi
+fi
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID"
+fi
+
+
+# set dependency paths
+
+
+ViashDebug "Running command: $(echo $VIASH_CMD)"
+cat << VIASHEOF | eval $VIASH_CMD
+set -e
+tempscript=\$(mktemp "$VIASH_META_TEMP_DIR/viash-run-generate_html-XXXXXX").sh
+function clean_up {
+  rm "\$tempscript"
+}
+function interrupt {
+  echo -e "\nCTRL-C Pressed..."
+  exit 1
+}
+trap clean_up EXIT
+trap interrupt INT SIGINT
+cat > "\$tempscript" << 'VIASHMAIN'
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT_DATA+x} ]; then echo "${VIASH_PAR_INPUT_DATA}" | sed "s#'#'\"'\"'#g;s#.*#par_input_data='&'#" ; else echo "# par_input_data="; fi )
+$( if [ ! -z ${VIASH_PAR_INPUT_STRUCTURE+x} ]; then echo "${VIASH_PAR_INPUT_STRUCTURE}" | sed "s#'#'\"'\"'#g;s#.*#par_input_structure='&'#" ; else echo "# par_input_structure="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_QC_REPORT+x} ]; then echo "${VIASH_PAR_OUTPUT_QC_REPORT}" | sed "s#'#'\"'\"'#g;s#.*#par_output_qc_report='&'#" ; else echo "# par_output_qc_report="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\"'\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\"'\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\"'\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\"'\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\"'\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\"'\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\"'\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\"'\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+ABSOLUTE_INPUT_DATA=\$(realpath \$par_input_data)
+ABSOLUTE_INPUT_STRUCTURE=\$(realpath \$par_input_structure)
+ABSOLUTE_OUTPUT=\$(realpath \$par_output_qc_report)
+
+cd /opt/incubator_ingestion_qc
+mkdir src/data
+
+echo "Compressing input data..."
+pnpm run compress_data "\$ABSOLUTE_INPUT_DATA" "src/data/dataset.ts"
+
+echo "Compressing report structure..."
+pnpm run compress_data "\$ABSOLUTE_INPUT_STRUCTURE" "src/data/report_structure.ts"
+
+echo "Generating HTML..."
+pnpm run build
+
+echo "Copying HTML to output directory..."
+cp dist/index.html "\$ABSOLUTE_OUTPUT"
+VIASHMAIN
+bash "\$tempscript" &
+wait "\$!"
+
+VIASHEOF
+
+
+if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
+  # strip viash automount from file paths
+  
+  if [ ! -z "$VIASH_PAR_INPUT_DATA" ]; then
+    VIASH_PAR_INPUT_DATA=$(ViashDockerStripAutomount "$VIASH_PAR_INPUT_DATA")
+  fi
+  if [ ! -z "$VIASH_PAR_INPUT_STRUCTURE" ]; then
+    VIASH_PAR_INPUT_STRUCTURE=$(ViashDockerStripAutomount "$VIASH_PAR_INPUT_STRUCTURE")
+  fi
+  if [ ! -z "$VIASH_PAR_OUTPUT_QC_REPORT" ]; then
+    VIASH_PAR_OUTPUT_QC_REPORT=$(ViashDockerStripAutomount "$VIASH_PAR_OUTPUT_QC_REPORT")
+  fi
+  if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
+    VIASH_META_RESOURCES_DIR=$(ViashDockerStripAutomount "$VIASH_META_RESOURCES_DIR")
+  fi
+  if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
+    VIASH_META_EXECUTABLE=$(ViashDockerStripAutomount "$VIASH_META_EXECUTABLE")
+  fi
+  if [ ! -z "$VIASH_META_CONFIG" ]; then
+    VIASH_META_CONFIG=$(ViashDockerStripAutomount "$VIASH_META_CONFIG")
+  fi
+  if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
+    VIASH_META_TEMP_DIR=$(ViashDockerStripAutomount "$VIASH_META_TEMP_DIR")
+  fi
+fi
+
+
+# check whether required files exist
+if [ ! -z "$VIASH_PAR_OUTPUT_QC_REPORT" ] && [ ! -e "$VIASH_PAR_OUTPUT_QC_REPORT" ]; then
+  ViashError "Output file '$VIASH_PAR_OUTPUT_QC_REPORT' does not exist."
+  exit 1
+fi
+
+
+exit 0
diff --git a/target/executable/ingestion_qc/generate_html/nextflow_labels.config b/target/executable/ingestion_qc/generate_html/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/executable/ingestion_qc/generate_html/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/.config.vsh.yaml b/target/nextflow/ingestion_qc/detect_ingestion_method/.config.vsh.yaml
new file mode 100644
index 0000000..81674ec
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/.config.vsh.yaml
@@ -0,0 +1,265 @@
+name: "detect_ingestion_method"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu file(s)"
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "The modality to use"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "string"
+    name: "--output_uns_ingestion_method"
+    description: "The .uns field in which to store the exprimental setup. Values stored\
+      \ are `cellranger_multi`, `xenium` or `cosmx`."
+    info: null
+    default:
+    - "ingestion_method"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "The output h5mu file, containing an .uns field with experiment description."
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Detects the ingestion method of a dataset.\nCurrently detects either\
+  \ 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other\
+  \ technologies upon request.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "sample_one.qc.h5mu"
+- type: "file"
+  path: "Lung5_Rep2_tiny.qc.h5mu"
+- type: "file"
+  path: "xenium_tiny.qc.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/detect_ingestion_method/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/ingestion_qc/detect_ingestion_method"
+  executable: "target/nextflow/ingestion_qc/detect_ingestion_method/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/main.nf b/target/nextflow/ingestion_qc/detect_ingestion_method/main.nf
new file mode 100644
index 0000000..00053e8
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/main.nf
@@ -0,0 +1,3998 @@
+// detect_ingestion_method build_main
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "detect_ingestion_method",
+  "namespace" : "ingestion_qc",
+  "version" : "build_main",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The input h5mu file(s)",
+          "example" : [
+            "path/to/file.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "The modality to use",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--output_uns_ingestion_method",
+          "description" : "The .uns field in which to store the exprimental setup. Values stored are `cellranger_multi`, `xenium` or `cosmx`.",
+          "default" : [
+            "ingestion_method"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The output h5mu file, containing an .uns field with experiment description.",
+          "example" : [
+            "path/to/file.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Detects the ingestion method of a dataset.\nCurrently detects either 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other technologies upon request.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/qc_sample_data/sample_one.qc.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/Lung5_Rep2_tiny.qc.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/xenium_tiny.qc.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "repositories" : [
+    {
+      "type" : "github",
+      "name" : "openpipeline",
+      "repo" : "openpipelines-bio/openpipeline",
+      "tag" : "2.1.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "lowdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "build_main",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/ingestion_qc/detect_ingestion_method/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/ingestion_qc/detect_ingestion_method",
+    "viash_version" : "0.9.4",
+    "git_commit" : "dfd68e55fafa27fc489f14bfab8ad04315f7f233",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_incubator",
+    "version" : "build_main",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "github",
+        "name" : "openpipeline",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'build_main'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import shutil
+import anndata as ad
+import h5py
+import sys
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_uns_ingestion_method': $( if [ ! -z ${VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD+x} ]; then echo "r'${VIASH_PAR_OUTPUT_UNS_INGESTION_METHOD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main(par):
+
+    # read h5mu file
+    with h5py.File(par["input"], "r") as file:
+        mod = file["mod"][par["modality"]]
+        uns = ad.experimental.read_elem(file["uns"])
+        mod_obs = ad.experimental.read_elem(mod["obs"])
+        mod_uns = ad.experimental.read_elem(mod["uns"])
+
+    # detect ingestion method
+    ingestion_methods = {
+            "cellranger_multi": "metrics_cellranger" in uns,
+            "xenium": all(key in mod_obs for key in ["segmentation_method", "nucleus_area"]),
+            "cosmx": "spatial" in mod_uns and "fov" in mod_obs
+        }
+
+    # make sure only one ingestion method is detected
+    detected_methods = [method for method, detected in ingestion_methods.items() if detected]
+    methods_count = len(detected_methods)
+
+    if methods_count == 1:
+        detected_method = detected_methods[0]
+        logger.info(f"Detected ingestion method {detected_method}")
+
+    elif methods_count == 0:
+        raise ValueError("No ingestion method detected")
+
+    else:
+        raise ValueError(f"Multiple ingestion methods detected: {', '.join(detected_methods)}")
+
+    # check if mod_uns already contains a different detected method
+    if mod_uns.get(par["output_uns_ingestion_method"], detected_method) != detected_method:
+        raise ValueError(f"Field .uns['{par['output_uns_ingestion_method']}'] already exists and contains different value \\`{mod_uns.get(par['output_uns_ingestion_method'])}\\` than detected method (\\`{detected_method}\\`).")
+
+    # copy input to output
+    shutil.copy(par["input"], par["output"])
+
+    if par["output_uns_ingestion_method"] not in mod_uns:
+        with h5py.File (par["output"], "r+") as out_file:
+            out_file["uns"][par["output_uns_ingestion_method"]] = detected_method
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline_incubator/ingestion_qc/detect_ingestion_method",
+    "tag" : "build_main"
+  },
+  "label" : [
+    "lowmem",
+    "lowdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow.config b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow.config
new file mode 100644
index 0000000..20a898a
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'ingestion_qc/detect_ingestion_method'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'build_main'
+  description = 'Detects the ingestion method of a dataset.\nCurrently detects either 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other technologies upon request.\n'
+  author = 'Dorien Roosen, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_labels.config b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_schema.json b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_schema.json
new file mode 100644
index 0000000..7971631
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/nextflow_schema.json
@@ -0,0 +1,71 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "detect_ingestion_method",
+  "description": "Detects the ingestion method of a dataset.\nCurrently detects either 10X CellRanger Multi, 10X Xenium or Nanostring CosMx, but can be extended to other technologies upon request.\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input h5mu file(s)",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "The modality to use",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_uns_ingestion_method": {
+          "type": "string",
+          "description": "The .uns field in which to store the exprimental setup",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"ingestion_method\"`. ",
+          "default": "ingestion_method"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "The output h5mu file, containing an .uns field with experiment description.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"path/to/file.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/nextflow/ingestion_qc/detect_ingestion_method/setup_logger.py b/target/nextflow/ingestion_qc/detect_ingestion_method/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/nextflow/ingestion_qc/detect_ingestion_method/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/nextflow/ingestion_qc/generate_html/.config.vsh.yaml b/target/nextflow/ingestion_qc/generate_html/.config.vsh.yaml
new file mode 100644
index 0000000..73c3291
--- /dev/null
+++ b/target/nextflow/ingestion_qc/generate_html/.config.vsh.yaml
@@ -0,0 +1,267 @@
+name: "generate_html"
+namespace: "ingestion_qc"
+version: "build_main"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input_data"
+    description: "The input JSON file containing the QC metrics"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input_structure"
+    description: "The input JSON file containing the structure of the data"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_qc_report"
+    description: "The output HTML report"
+    info: null
+    example:
+    - "path/to/file.html"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Generate an HTML report from the QC metrics"
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+- type: "file"
+  path: "sc_dataset.json"
+- type: "file"
+  path: "sc_report_structure.json"
+- type: "file"
+  path: "xenium_dataset.json"
+- type: "file"
+  path: "xenium_report_structure.json"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "node:latest"
+  target_registry: "images.viash-hub.com"
+  target_tag: "build_main"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "docker"
+    run:
+    - "npm install -g pnpm@latest-10 \\\n&& cd /opt && git clone https://github.com/openpipelines-bio/incubator_ingestion_qc.git\
+      \ \\\n&& cd incubator_ingestion_qc && pnpm install \\\n&& true\n"
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/generate_html/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/ingestion_qc/generate_html"
+  executable: "target/nextflow/ingestion_qc/generate_html/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/nextflow/ingestion_qc/generate_html/main.nf b/target/nextflow/ingestion_qc/generate_html/main.nf
new file mode 100644
index 0000000..e401b57
--- /dev/null
+++ b/target/nextflow/ingestion_qc/generate_html/main.nf
@@ -0,0 +1,3966 @@
+// generate_html build_main
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "generate_html",
+  "namespace" : "ingestion_qc",
+  "version" : "build_main",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input_data",
+          "description" : "The input JSON file containing the QC metrics",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input_structure",
+          "description" : "The input JSON file containing the structure of the data",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_qc_report",
+          "description" : "The output HTML report",
+          "example" : [
+            "path/to/file.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Generate an HTML report from the QC metrics",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/qc_sample_data/sc_dataset.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/qc_sample_data/sc_report_structure.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/xenium_dataset.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/xenium_report_structure.json"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "repositories" : [
+    {
+      "type" : "github",
+      "name" : "openpipeline",
+      "repo" : "openpipelines-bio/openpipeline",
+      "tag" : "2.1.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "lowdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "node:latest",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "build_main",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "docker",
+          "run" : [
+            "npm install -g pnpm@latest-10 \\\\\n&& cd /opt && git clone https://github.com/openpipelines-bio/incubator_ingestion_qc.git \\\\\n&& cd incubator_ingestion_qc && pnpm install \\\\\n&& true\n"
+          ]
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/ingestion_qc/generate_html/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/ingestion_qc/generate_html",
+    "viash_version" : "0.9.4",
+    "git_commit" : "dfd68e55fafa27fc489f14bfab8ad04315f7f233",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_incubator",
+    "version" : "build_main",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "github",
+        "name" : "openpipeline",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'build_main'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT_DATA+x} ]; then echo "${VIASH_PAR_INPUT_DATA}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input_data='&'#" ; else echo "# par_input_data="; fi )
+$( if [ ! -z ${VIASH_PAR_INPUT_STRUCTURE+x} ]; then echo "${VIASH_PAR_INPUT_STRUCTURE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input_structure='&'#" ; else echo "# par_input_structure="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_QC_REPORT+x} ]; then echo "${VIASH_PAR_OUTPUT_QC_REPORT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output_qc_report='&'#" ; else echo "# par_output_qc_report="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+ABSOLUTE_INPUT_DATA=\\$(realpath \\$par_input_data)
+ABSOLUTE_INPUT_STRUCTURE=\\$(realpath \\$par_input_structure)
+ABSOLUTE_OUTPUT=\\$(realpath \\$par_output_qc_report)
+
+cd /opt/incubator_ingestion_qc
+mkdir src/data
+
+echo "Compressing input data..."
+pnpm run compress_data "\\$ABSOLUTE_INPUT_DATA" "src/data/dataset.ts"
+
+echo "Compressing report structure..."
+pnpm run compress_data "\\$ABSOLUTE_INPUT_STRUCTURE" "src/data/report_structure.ts"
+
+echo "Generating HTML..."
+pnpm run build
+
+echo "Copying HTML to output directory..."
+cp dist/index.html "\\$ABSOLUTE_OUTPUT"
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline_incubator/ingestion_qc/generate_html",
+    "tag" : "build_main"
+  },
+  "label" : [
+    "lowmem",
+    "lowdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/nextflow/ingestion_qc/generate_html/nextflow.config b/target/nextflow/ingestion_qc/generate_html/nextflow.config
new file mode 100644
index 0000000..b0e6b11
--- /dev/null
+++ b/target/nextflow/ingestion_qc/generate_html/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'ingestion_qc/generate_html'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'build_main'
+  description = 'Generate an HTML report from the QC metrics'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/nextflow/ingestion_qc/generate_html/nextflow_labels.config b/target/nextflow/ingestion_qc/generate_html/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/nextflow/ingestion_qc/generate_html/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/ingestion_qc/generate_html/nextflow_schema.json b/target/nextflow/ingestion_qc/generate_html/nextflow_schema.json
new file mode 100644
index 0000000..e21af43
--- /dev/null
+++ b/target/nextflow/ingestion_qc/generate_html/nextflow_schema.json
@@ -0,0 +1,66 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "generate_html",
+  "description": "Generate an HTML report from the QC metrics",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input_data": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input JSON file containing the QC metrics",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file.json\"`. "
+        },
+        "input_structure": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input JSON file containing the structure of the data",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file.json\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_qc_report": {
+          "type": "string",
+          "format": "path",
+          "description": "The output HTML report",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_qc_report.html\"`, direction: `output`, example: `\"path/to/file.html\"`. ",
+          "default": "$id.$key.output_qc_report.html"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/nextflow/workflows/generate_qc_report/.config.vsh.yaml b/target/nextflow/workflows/generate_qc_report/.config.vsh.yaml
new file mode 100644
index 0000000..26b192c
--- /dev/null
+++ b/target/nextflow/workflows/generate_qc_report/.config.vsh.yaml
@@ -0,0 +1,395 @@
+name: "generate_qc_report"
+namespace: "workflows"
+version: "build_main"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "The sample IDs to include in the report. If not provided,\nthe sample\
+      \ IDs will be extracted from the h5mu files.\n"
+    info: null
+    example:
+    - "sample1"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu files."
+    info: null
+    example:
+    - "path/to/file1.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ingestion_method"
+    info: null
+    required: true
+    choices:
+    - "cellranger_multi"
+    - "xenium"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--sample_metadata"
+    description: "The sample metadata file corresponding to .obs fields in the h5mu\
+      \ input files, to be used for grouping in the report.\n"
+    info: null
+    example:
+    - "path/to/file.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_samples_per_report"
+    description: "The maximum number of samples to be included per report. \nMultiple\
+      \ reports will be generated (with samples equally divided over all reports)\
+      \ if number of input samples exceeds this threshold.\n"
+    info: null
+    default:
+    - 20
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: "The column name in the .var h5mu files that contains the gene names.\
+      \ If not provided, .var_names will be used.\n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_metadata"
+    description: "The metadata keys in the h5mu .obs to include in the report."
+    info: null
+    example:
+    - "donor_id"
+    - "cell_type"
+    - "batch"
+    - "condition"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "QC options"
+  arguments:
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    default:
+    - "mitochondrial"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    default:
+    - "ribosomal"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Cellbender options"
+  arguments:
+  - type: "boolean"
+    name: "--run_cellbender"
+    description: "Whether to run cellbender or not."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--cellbender_epochs"
+    description: "Number of epochs to train cellbender."
+    info: null
+    default:
+    - 150
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_qc_report"
+    description: "The output HTML report"
+    info: null
+    example:
+    - "path/to/file.html"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_processed_h5mu"
+    description: "Folder containing the processed h5mu files."
+    info: null
+    default:
+    - "qc_h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Run the ingestion QC report generation"
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_no_cellbender"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_with_cellbender"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+dependencies:
+- name: "metadata/add_id"
+  repository:
+    type: "github"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+- name: "workflows/qc/qc"
+  repository:
+    type: "github"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+- name: "correction/cellbender_remove_background"
+  alias: "cellbender"
+  repository:
+    type: "github"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+- name: "ingestion_qc/h5mu_to_qc_json"
+  repository:
+    type: "local"
+- name: "ingestion_qc/generate_html"
+  repository:
+    type: "local"
+- name: "move_files_to_directory"
+  repository:
+    type: "vsh"
+    repo: "craftbox"
+    tag: "v0.2.0"
+repositories:
+- type: "github"
+  name: "openpipeline"
+  repo: "openpipelines-bio/openpipeline"
+  tag: "2.1.2"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/generate_qc_report/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/generate_qc_report"
+  executable: "target/nextflow/workflows/generate_qc_report/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "dfd68e55fafa27fc489f14bfab8ad04315f7f233"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+  dependencies:
+  - "target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id"
+  - "target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc"
+  - "target/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background"
+  - "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json"
+  - "target/nextflow/ingestion_qc/generate_html"
+  - "target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory"
+package_config:
+  name: "openpipeline_incubator"
+  version: "build_main"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "github"
+    name: "openpipeline"
+    repo: "openpipelines-bio/openpipeline"
+    tag: "2.1.2"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'build_main'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_incubator"
+    docker_registry: "ghcr.io"
diff --git a/target/nextflow/workflows/generate_qc_report/main.nf b/target/nextflow/workflows/generate_qc_report/main.nf
new file mode 100644
index 0000000..02e402a
--- /dev/null
+++ b/target/nextflow/workflows/generate_qc_report/main.nf
@@ -0,0 +1,3930 @@
+// generate_qc_report build_main
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "generate_qc_report",
+  "namespace" : "workflows",
+  "version" : "build_main",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "The sample IDs to include in the report. If not provided,\nthe sample IDs will be extracted from the h5mu files.\n",
+          "example" : [
+            "sample1"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The input h5mu files.",
+          "example" : [
+            "path/to/file1.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ingestion_method",
+          "required" : true,
+          "choices" : [
+            "cellranger_multi",
+            "xenium"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--sample_metadata",
+          "description" : "The sample metadata file corresponding to .obs fields in the h5mu input files, to be used for grouping in the report.\n",
+          "example" : [
+            "path/to/file.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_samples_per_report",
+          "description" : "The maximum number of samples to be included per report. \nMultiple reports will be generated (with samples equally divided over all reports) if number of input samples exceeds this threshold.\n",
+          "default" : [
+            20
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : "The column name in the .var h5mu files that contains the gene names. If not provided, .var_names will be used.\n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_metadata",
+          "description" : "The metadata keys in the h5mu .obs to include in the report.",
+          "example" : [
+            "donor_id",
+            "cell_type",
+            "batch",
+            "condition"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "default" : [
+            "mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "default" : [
+            "ribosomal"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Cellbender options",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--run_cellbender",
+          "description" : "Whether to run cellbender or not.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--cellbender_epochs",
+          "description" : "Number of epochs to train cellbender.",
+          "default" : [
+            150
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_qc_report",
+          "description" : "The output HTML report",
+          "example" : [
+            "path/to/file.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_processed_h5mu",
+          "description" : "Folder containing the processed h5mu files.",
+          "default" : [
+            "qc_h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Run the ingestion QC report generation",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_no_cellbender"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_with_cellbender"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/add_id",
+      "repository" : {
+        "type" : "github",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      }
+    },
+    {
+      "name" : "workflows/qc/qc",
+      "repository" : {
+        "type" : "github",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      }
+    },
+    {
+      "name" : "correction/cellbender_remove_background",
+      "alias" : "cellbender",
+      "repository" : {
+        "type" : "github",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      }
+    },
+    {
+      "name" : "ingestion_qc/h5mu_to_qc_json",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "ingestion_qc/generate_html",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "move_files_to_directory",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    }
+  ],
+  "repositories" : [
+    {
+      "type" : "github",
+      "name" : "openpipeline",
+      "repo" : "openpipelines-bio/openpipeline",
+      "tag" : "2.1.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/generate_qc_report/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/generate_qc_report",
+    "viash_version" : "0.9.4",
+    "git_commit" : "dfd68e55fafa27fc489f14bfab8ad04315f7f233",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_incubator",
+    "version" : "build_main",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "github",
+        "name" : "openpipeline",
+        "repo" : "openpipelines-bio/openpipeline",
+        "tag" : "2.1.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'build_main'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_incubator",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { add_id } from "${meta.root_dir}/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/metadata/add_id/main.nf"
+include { qc } from "${meta.root_dir}/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/workflows/qc/qc/main.nf"
+include { cellbender_remove_background as cellbender_viashalias } from "${meta.root_dir}/dependencies/github/openpipelines-bio/openpipeline/2.1.2/nextflow/correction/cellbender_remove_background/main.nf"
+cellbender = cellbender_viashalias.run(key: "cellbender")
+include { h5mu_to_qc_json } from "${meta.resources_dir}/../../../_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf"
+include { generate_html } from "${meta.resources_dir}/../../../nextflow/ingestion_qc/generate_html/main.nf"
+include { move_files_to_directory } from "${meta.root_dir}/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take: input_ch
+  main:
+  qc_ch = input_ch
+    // store join id
+    | map { id, state ->
+      [id, state + [_meta: [join_id: id]]]
+    }
+
+    // add sample ids to each state
+    | add_id.run(
+      fromState: [
+        input_id: "id", 
+        input: "input"
+      ],
+      args: [
+        obs_output: "sample_id"
+      ],
+      toState: [ "input": "output" ]
+    )
+
+    // run cellbender
+    | cellbender.run(
+      runIf: {id, state -> state.run_cellbender},
+      fromState: [
+        id: "id",
+        input: "input",
+        epochs: "cellbender_epochs",
+      ],
+      args: [
+        obs_background_fraction: "cellbender_background_fraction",
+        obs_cell_probability: "cellbender_cell_probability",
+        obs_droplet_efficiency: "cellbender_droplet_efficiency",
+        obs_cell_size: "cellbender_cell_size",
+      ],
+      toState: ["input": "output"]
+    )
+
+    // run qc on each sample
+    | qc.run(
+      fromState: [
+        id: "id",
+        input: "input",
+        var_gene_names: "var_gene_names"
+      ],
+      args: [
+        var_name_mitochondrial_genes: "mitochondrial",
+        var_name_ribosomal_genes: "ribosomal",
+        output_obs_num_nonzero_vars: "num_nonzero_vars",
+        output_obs_total_counts_vars: "total_counts"
+      ],
+      toState: { id, output, state ->
+        def keysToRemove = ["var_gene_names", "var_name_mitochondrial_genes", "var_name_ribosomal_genes", "run_cellbender", "cellbender_epochs"]
+        def newState = state.findAll{it.key !in keysToRemove}
+        newState + ["input": output.output]
+      }
+    )
+
+    | joinStates { ids, states ->
+      def newId = "qc_data"
+      // gather keys with unique values across states that should be combined
+      def new_state_non_unique_values = [
+        input: states.collect{it.input},
+        join_ids: states.collect{it._meta.join_id},
+        _meta: [join_id: ids[0]]
+      ]
+      // gather keys from different states
+      def all_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+        }.minus(["output", "id", "input", "_meta"])
+      // Create the new state from the keys, values should be the same across samples
+      def new_state = all_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across samples. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            // take the unique value from the set (there is only one)
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+        }
+      def data_state = new_state_non_unique_values + new_state
+      [ newId, data_state ]
+    }
+
+  processed_files_ch = qc_ch
+
+    // move all processed h5mu files to the same folder
+    | move_files_to_directory.run(
+      fromState: [
+        input: "input",
+        output: "output_processed_h5mu"
+      ],
+      toState: [ "output_processed_h5mu": "output" ]
+    )
+    | setState(["output_processed_h5mu"])
+
+  report_ch = qc_ch
+    // group the processed samples to generate one or multiple reports
+    | flatMap { id, state ->
+
+        // calculate number of reports to be generated and number of samples per report
+        def totalInputs = state.input.size()
+        def maxSamplesPerGroup = state.max_samples_per_report
+        def numGroups = Math.max(1, Math.ceil(totalInputs / maxSamplesPerGroup) as Integer)
+        def baseSamplesPerGroup = totalInputs.intdiv(numGroups)
+        def remainder = totalInputs % numGroups
+
+        println "Dividing ${totalInputs} sample(s) over ${numGroups} report(s) (max ${maxSamplesPerGroup} per report)"
+        
+        // sort inputs to make grouping deterministic
+        def inputs = []
+        for (int i = 0; i < state.input.size(); i++) {
+            inputs << [input: state.input[i], _meta: [join_id: state.join_ids[i]]]
+        }
+
+        def sortedInputs = inputs.sort { it._meta.join_id }
+  
+        def groups = []
+        def itemIndex = 0
+
+        // create one channel per report
+        (0..<numGroups).each { groupNum ->
+            def samplesInGroup = baseSamplesPerGroup + (groupNum < remainder ? 1 : 0)
+            def groupItems = sortedInputs[itemIndex..<(itemIndex + samplesInGroup)]
+            
+            def newId = "combined_${groupNum + 1}_of_${numGroups}"
+            def newState = state.clone()  // Copy all the original state
+            
+            // Override the input and _meta with the grouped items
+            newState.input = groupItems.collect { it.input }
+            newState._meta = groupItems[0]._meta
+            
+            println "Group ${groupNum + 1}: ${samplesInGroup} samples - ${newState._meta}"
+            
+            groups << [newId, newState]
+            itemIndex += samplesInGroup
+        }
+        
+        return groups
+    }
+
+    // Set aside output for QC report instructions
+    | map { id, state -> 
+      def new_state = state + ["output_reporting_json": "reporting_json.json"]
+      [id, new_state]
+    }
+
+    // Set report filter settings
+    | map { id, state -> 
+      def conditionalValues = [
+        "cellranger_multi": [
+          "min_total_counts": 10,
+          "min_num_nonzero_vars": 10
+        ],
+        "xenium": [
+          "min_total_counts": 10, 
+          "min_num_nonzero_vars": 1
+        ]
+      ]
+      
+      def method = state.ingestion_method
+      def additionalParams = conditionalValues[method]
+      
+      [ id, state + additionalParams ]
+    }
+
+    | view {"After setting filters: $it"}
+
+    // generate qc json
+    | h5mu_to_qc_json.run(
+      fromState: [
+        input: "input",
+        ingestion_method: "ingestion_method",
+        obs_metadata: "obs_metadata",
+        min_total_counts: "min_total_counts",
+        min_num_nonzero_vars: "min_num_nonzero_vars"
+      ],
+      args: [
+        obs_sample_id: "sample_id",
+        obs_total_counts: "total_counts",
+        obs_num_nonzero_vars: "num_nonzero_vars",
+        obs_fraction_mitochondrial: "fraction_mitochondrial",
+        obs_fraction_ribosomal: "fraction_ribosomal",
+      ],
+      toState: [
+        output: "output",
+        output_reporting_json: "output_reporting_json"
+      ]
+    )
+
+    // generate html report
+    | generate_html.run(
+      fromState: [ 
+        input_data: "output",
+        input_structure: "output_reporting_json"
+      ],
+      toState: [
+        output_qc_report: "output_qc_report"
+      ]
+    )
+
+    // collect the reports into a single channel
+    | joinStates { ids, states ->
+      def newId = "qc_report"
+      def report_state = [
+        output_qc_report: states.collect{it.output_qc_report},
+        _meta: states[0]._meta
+      ]
+      [ newId, report_state ]
+    }
+    
+  output_ch = report_ch.mix(processed_files_ch)
+
+    | joinStates { ids, states ->
+
+      assert states.size() == 2, "Expected 2 states, but got ${states.size()}"
+      assert ids.contains('qc_report'), "Expected one channel to have the id `qc_report`, but got ${ids}"
+      assert ids.contains('qc_data'), "Expected one channel to have the id `qc_data`, but got ${ids}"
+
+      def newId = "combined"
+      def combined_state = states[0] + states [1]
+
+      [ newId, combined_state ]
+    }
+
+  emit: output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/nextflow/workflows/generate_qc_report/nextflow.config b/target/nextflow/workflows/generate_qc_report/nextflow.config
new file mode 100644
index 0000000..65e419b
--- /dev/null
+++ b/target/nextflow/workflows/generate_qc_report/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/generate_qc_report'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'build_main'
+  description = 'Run the ingestion QC report generation'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/nextflow/workflows/generate_qc_report/nextflow_labels.config b/target/nextflow/workflows/generate_qc_report/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/nextflow/workflows/generate_qc_report/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/workflows/generate_qc_report/nextflow_schema.json b/target/nextflow/workflows/generate_qc_report/nextflow_schema.json
new file mode 100644
index 0000000..1d8c245
--- /dev/null
+++ b/target/nextflow/workflows/generate_qc_report/nextflow_schema.json
@@ -0,0 +1,162 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "generate_qc_report",
+  "description": "Run the ingestion QC report generation",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "The sample IDs to include in the report",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"sample1\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input h5mu files.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file1.h5mu\"`. "
+        },
+        "ingestion_method": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, required, choices: ``cellranger_multi`, `xenium``. ",
+          "enum": [
+            "cellranger_multi",
+            "xenium"
+          ]
+        },
+        "sample_metadata": {
+          "type": "string",
+          "format": "path",
+          "description": "The sample metadata file corresponding to .obs fields in the h5mu input files, to be used for grouping in the report.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"path/to/file.csv\"`. "
+        },
+        "max_samples_per_report": {
+          "type": "integer",
+          "description": "The maximum number of samples to be included per report",
+          "help_text": "Type: `integer`, multiple: `False`, default: `20`. ",
+          "default": 20
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_qc_report": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The output HTML report",
+          "help_text": "Type: `file`, multiple: `True`, required, default: `\"$id.$key.output_qc_report_*.html\"`, direction: `output`, example: `[\"path/to/file.html\"]`. ",
+          "default": "$id.$key.output_qc_report_*.html"
+        },
+        "output_processed_h5mu": {
+          "type": "string",
+          "format": "path",
+          "description": "Folder containing the processed h5mu files.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"qc_h5mu\"`, direction: `output`. ",
+          "default": "qc_h5mu"
+        }
+      }
+    },
+    "options": {
+      "title": "Options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_gene_names": {
+          "type": "string",
+          "description": "The column name in the .var h5mu files that contains the gene names",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gene_symbol\"`. "
+        },
+        "obs_metadata": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The metadata keys in the h5mu .obs to include in the report.",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"donor_id\";\"cell_type\";\"batch\";\"condition\"]`. "
+        }
+      }
+    },
+    "qc options": {
+      "title": "QC options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"mitochondrial\"`. ",
+          "default": "mitochondrial"
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"ribosomal\"`. ",
+          "default": "ribosomal"
+        }
+      }
+    },
+    "cellbender options": {
+      "title": "Cellbender options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "run_cellbender": {
+          "type": "boolean",
+          "description": "Whether to run cellbender or not.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "cellbender_epochs": {
+          "type": "integer",
+          "description": "Number of epochs to train cellbender.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `150`. ",
+          "default": 150
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/options"
+    },
+    {
+      "$ref": "#/$defs/qc options"
+    },
+    {
+      "$ref": "#/$defs/cellbender options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}