fix(modules): address awgymer review comments on PR #139 (strict nf-core compliance)

.github/workflows/nf-test.yml

@@ -64,7 +64,7 @@ jobs:
     runs-on: # use self-hosted runners
       - runs-on=${{ github.run_id }}-nf-test
       - runner=4cpu-linux-x64
-      - disk=large
+      - volume=80gb
     strategy:
       fail-fast: false
       matrix:

bin/spatialdata_merge.py

@@ -6,8 +6,6 @@
 import os
 import shutil
 
-import spatialdata
-
 
 def parse_args():
     """Parse command-line arguments."""

bin/spatialdata_meta.py

@@ -7,7 +7,6 @@
 
 import pandas as pd
 import spatialdata as sd
-import zarr
 
 # Fix zarr v3 + anndata + numcodecs incompatibility:
 # anndata's string writer passes numcodecs.VLenUTF8 to zarr.Group.create_array,

bin/spatialdata_write.py

@@ -5,7 +5,6 @@
 import sys
 
 import pandas as pd
-import spatialdata
 from spatialdata_io import xenium
 
 # Fix zarr v3 + anndata + numcodecs incompatibility:

conf/modules.config

@@ -20,10 +20,6 @@ process {
 
     // ---------------------------- multiqc ---------------------------------------------------
 
-    withName: 'MULTIQC|MULTIQC_PRE_XR_RUN|MULTIQC_POST_XR_RUN' {
-        errorStrategy = 'ignore'
-    }
-
     withName: MULTIQC {
         ext.args   = { params.multiqc_title ? "--title \"${params.multiqc_title}\"" : '' }
         publishDir = [
@@ -72,13 +68,23 @@ process {
             path: "${params.outdir}/${params.mode}/xeniumranger/resegment",
             mode: params.publish_dir_mode,
         ]
+        ext.args = {[
+            // Disable boundary/interior stain when the param is falsy; keep tool default when truthy.
+            !params.boundary_stain                ? "--boundary-stain=disable"                       : "",
+            !params.interior_stain                ? "--interior-stain=disable"                       : "",
+            params.expansion_distance != null ? "--expansion-distance=${params.expansion_distance}" : "",
+            params.dapi_filter        != null ? "--dapi-filter=${params.dapi_filter}"               : "",
+        ].join(' ').trim()}
     }
 
     withName: XENIUMRANGER_IMPORT_SEGMENTATION {
         publishDir = [
             path: "${params.outdir}/${params.mode}/xeniumranger/import_segementation",
             mode: params.publish_dir_mode,
         ]
+        ext.args = {[
+            params.expansion_distance != null ? "--expansion-distance=${params.expansion_distance}" : "",
+        ].join(' ').trim()}
     }
 
     // ---------------------------- proseg ---------------------------------------------------
@@ -88,7 +94,9 @@ process {
             path: "${params.outdir}/${params.mode}/proseg/preset",
             mode: params.publish_dir_mode,
         ]
-        ext.format = { params.format ?: 'xenium' }
+        ext.args = {[
+            params.format != null ? "--${params.format}" : "",
+        ].join(' ').trim()}
     }
 
     withName: PROSEG2BAYSOR {
@@ -173,19 +181,24 @@ process {
             path: { "${params.outdir}/${params.mode}/segger/create_dataset" },
             mode: params.publish_dir_mode,
         ]
-        ext.args   = { "--tile-width ${params.tile_width} --tile-height ${params.tile_height}" }
-        ext.format = { params.format ?: 'xenium' }
+        ext.args = {[
+            params.format      != null ? "--sample-type ${params.format}"      : "",
+            params.tile_width  != null ? "--tile-width ${params.tile_width}"   : "",
+            params.tile_height != null ? "--tile-height ${params.tile_height}" : "",
+        ].join(' ').trim()}
     }
 
     withName: SEGGER_TRAIN {
         publishDir = [
             path: { "${params.outdir}/${params.mode}/segger/train" },
             mode: params.publish_dir_mode,
         ]
-        ext.devices = params.devices
-        ext.args = { "--batch_size ${params.batch_size_train} --max_epochs ${params.max_epochs} --num_workers ${params.segger_num_workers}" }
-        ext.args2   = { "--init_emb 8 --hidden_channels 32 --num_tx_tokens 10000 --out_channels 8 --heads 2 --num_mid_layers 2 --strategy auto --precision bf16-mixed" }
-        maxForks = params.restrict_concurrency ? 1 : 0
+        ext.args = {[
+            "--init_emb 8 --hidden_channels 32 --num_tx_tokens 10000 --out_channels 8 --heads 2 --num_mid_layers 2 --strategy auto --precision bf16-mixed",
+            params.batch_size_train   != null ? "--batch_size ${params.batch_size_train}"    : "",
+            params.max_epochs         != null ? "--max_epochs ${params.max_epochs}"          : "",
+            params.segger_num_workers != null ? "--num_workers ${params.segger_num_workers}" : "",
+        ].join(' ').trim()}
     }
 
     withName: SEGGER_PREDICT {
@@ -195,7 +208,11 @@ process {
             // Skip partitioned parquet dirs (Hive-style) that S3 copy can't handle
             saveAs: { filename -> filename.contains('transcripts_df.parquet') ? null : filename },
         ]
-        ext.args   = { "--batch-size ${params.batch_size_predict} --cc-analysis ${params.cc_analysis} --knn-method ${params.segger_knn_method}" }
+        ext.args = {[
+            params.batch_size_predict != null ? "--batch-size ${params.batch_size_predict}" : "",
+            params.cc_analysis        != null ? "--cc-analysis ${params.cc_analysis}"       : "",
+            params.segger_knn_method  != null ? "--knn-method ${params.segger_knn_method}"  : "",
+        ].join(' ').trim()}
     }
 
     // ---------------------------- ficture ------------------------------------------

main.nf

@@ -105,6 +105,7 @@ workflow {
         params.help,
         params.help_full,
         params.show_hidden,
+        params.format,
         params.gene_panel,
         params.gene_synonyms,
         params.image_seg_methods,

modules.json

@@ -56,8 +56,7 @@
                     "xeniumranger/import-segmentation": {
                         "branch": "master",
                         "git_sha": "39365e944e936511e33b993cdd978e0f12adac9a",
-                        "installed_by": ["modules"],
-                        "patch": "modules/nf-core/xeniumranger/import-segmentation/xeniumranger-import-segmentation.diff"
+                        "installed_by": ["modules"]
                     },
                     "xeniumranger/relabel": {
                         "branch": "master",
@@ -67,8 +66,7 @@
                     "xeniumranger/resegment": {
                         "branch": "master",
                         "git_sha": "39365e944e936511e33b993cdd978e0f12adac9a",
-                        "installed_by": ["modules"],
-                        "patch": "modules/nf-core/xeniumranger/resegment/xeniumranger-resegment.diff"
+                        "installed_by": ["modules"]
                     }
                 }
             },

modules/local/proseg/preset/main.nf

@@ -23,18 +23,12 @@ process PROSEG {
 
     def args = task.ext.args ?: ''
     prefix = task.ext.prefix ?: "${meta.id}"
-    def format = task.ext.format ?: 'xenium'
-
-    // check for platform values
-    if (!(format in ['xenium', 'cosmx', 'merscope'])) {
-        error("${format} is an invalid platform type. Please specify xenium, cosmx, or merscope")
-    }
 
     """
     mkdir -p ${prefix}
 
     proseg \\
-        --${format} \\
+        ${args} \\
         ${transcripts} \\
         --nthreads ${task.cpus} \\
         --output-expected-counts ${prefix}/expected-counts.csv.gz \\
@@ -45,8 +39,7 @@ process PROSEG {
         --output-cell-polygons ${prefix}/cell-polygons.geojson.gz \\
         --output-cell-polygon-layers ${prefix}/cell-polygons-layers.geojson.gz \\
         --output-union-cell-polygons ${prefix}/union-cell-polygons.geojson.gz \\
-        --output-spatialdata ${prefix}/proseg-output.zarr \\
-        ${args}
+        --output-spatialdata ${prefix}/proseg-output.zarr
     """
 
     stub:

modules/local/segger/create_dataset/main.nf

@@ -1,7 +1,6 @@
 process SEGGER_CREATE_DATASET {
     tag "${meta.id}"
     label 'process_xl'
-    maxForks params.restrict_concurrency ? 1 : 0
 
     container "quay.io/dongzehe/segger:1.0.14"
 
@@ -23,12 +22,6 @@ process SEGGER_CREATE_DATASET {
 
     def args = task.ext.args ?: ''
     prefix = task.ext.prefix ?: "${meta.id}"
-    def format = task.ext.format ?: 'xenium'
-
-    // check for platform values
-    if (!(format in ['xenium'])) {
-        error("${format} is an invalid platform type.")
-    }
 
     """
     export NUMBA_CACHE_DIR=\$PWD/.numba_cache
@@ -37,7 +30,6 @@ process SEGGER_CREATE_DATASET {
     segger_create_dataset.py \\
         --bundle-dir ${base_dir} \\
         --output-dir ${prefix} \\
-        --sample-type ${format} \\
         --n-workers ${task.cpus} \\
         ${args}
     """

modules/local/segger/train/main.nf

@@ -22,14 +22,11 @@ process SEGGER_TRAIN {
     }
 
     def args = task.ext.args ?: ''
-    def args2 = task.ext.args2 ?: ''
     def script_path = "/workspace/segger_dev/src/segger/cli/train_model.py"
     prefix = task.ext.prefix ?: "${meta.id}"
-    // Scale GPU count with retries: 4 → 8 (capped at params.devices)
     def gpu_count = 2 * task.attempt
     def cuda_visible = gpu_count == 1 ? "export CUDA_VISIBLE_DEVICES=0" : ""
     def accelerator = task.accelerator ? 'gpu' : 'auto'
-    def num_devices = task.devices ?: 0
 
     """
     # Set numba cache directory to avoid caching issues in container
@@ -38,7 +35,7 @@ process SEGGER_TRAIN {
 
     # GPU detection logging
     echo "=== GPU Detection (SEGGER_TRAIN) ==="
-    echo "Requested devices: ${gpu_count} (attempt ${task.attempt}, max ${num_devices})"
+    echo "Requested devices: ${gpu_count} (attempt ${task.attempt})"
     echo "Accelerator: ${accelerator}"
     nvidia-smi 2>/dev/null && echo "GPU available: yes" || echo "GPU available: no (nvidia-smi failed)"
     python3 -c "import torch; print(f'PyTorch CUDA available: {torch.cuda.is_available()}'); print(f'CUDA device count: {torch.cuda.device_count()}')" 2>/dev/null || echo "PyTorch CUDA check failed"
@@ -51,8 +48,7 @@ process SEGGER_TRAIN {
         --sample_tag ${prefix} \\
         --devices ${gpu_count} \\
         --accelerator ${accelerator} \\
-        ${args} \\
-        ${args2}
+        ${args}
     """
 
     stub:

modules/local/spatialdata/merge/main.nf

@@ -12,7 +12,7 @@ process SPATIALDATA_MERGE {
 
     output:
     tuple val(meta), path("spatialdata/${prefix}/${outputfolder}"), emit: merged_bundle
-    tuple val("${task.process}"), val('spatialdata'), eval('python3 -c "import spatialdata; print(spatialdata.__version__)"'), topic: versions, emit: versions_spatialdata
+    tuple val("${task.process}"), val('spatialdata'), eval("pip show spatialdata | sed -n 's/^Version: //p'"), topic: versions, emit: versions_spatialdata
 
     when:
     task.ext.when == null || task.ext.when

modules/local/spatialdata/meta/main.nf

@@ -12,7 +12,7 @@ process SPATIALDATA_META {
 
     output:
     tuple val(meta), path("spatialdata/${prefix}/${outputfolder}"), emit: metadata
-    tuple val("${task.process}"), val('spatialdata'), eval('python3 -c "import spatialdata; print(spatialdata.__version__)"'), topic: versions, emit: versions_spatialdata
+    tuple val("${task.process}"), val('spatialdata'), eval("pip show spatialdata | sed -n 's/^Version: //p'"), topic: versions, emit: versions_spatialdata
 
     when:
     task.ext.when == null || task.ext.when

modules/local/spatialdata/write/main.nf

@@ -14,7 +14,7 @@ process SPATIALDATA_WRITE {
 
     output:
     tuple val(meta), path("spatialdata/${prefix}/${outputfolder}"), emit: spatialdata
-    tuple val("${task.process}"), val('spatialdata'), eval('python3 -c "import spatialdata; print(spatialdata.__version__)"'), topic: versions, emit: versions_spatialdata
+    tuple val("${task.process}"), val('spatialdata'), eval("pip show spatialdata | sed -n 's/^Version: //p'"), topic: versions, emit: versions_spatialdata
 
     when:
     task.ext.when == null || task.ext.when

modules/local/utility/downscale_morphology/main.nf

@@ -30,8 +30,8 @@ process DOWNSCALE_MORPHOLOGY {
     tuple val(meta), path("${prefix}/downscaled.tif"), emit: downscaled
     tuple val(meta), path("${prefix}/scale_info.json"), emit: scale_info
     tuple val("${task.process}"), val('python'), eval("python3 --version | sed 's/Python //'"), topic: versions, emit: versions_python
-    tuple val("${task.process}"), val('tifffile'), eval('python3 -c "import tifffile; print(tifffile.__version__)"'), topic: versions, emit: versions_tifffile
-    tuple val("${task.process}"), val('scikit-image'), eval('python3 -c "import skimage; print(skimage.__version__)"'), topic: versions, emit: versions_skimage
+    tuple val("${task.process}"), val('tifffile'), eval("pip show tifffile 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_tifffile
+    tuple val("${task.process}"), val('scikit-image'), eval("pip show scikit-image 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_skimage
 
     when:
     task.ext.when == null || task.ext.when

modules/local/utility/parquet_to_csv/main.nf

@@ -12,7 +12,7 @@ process PARQUET_TO_CSV {
 
     output:
     tuple val(meta), path("${prefix}/*.csv*"), emit: transcripts_csv
-    tuple val("${task.process}"), val('pyarrow'), eval('python3 -c "import pyarrow; print(pyarrow.__version__)"'), topic: versions, emit: versions_pyarrow
+    tuple val("${task.process}"), val('pyarrow'), eval("pip show pyarrow 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_pyarrow
 
     when:
     task.ext.when == null || task.ext.when

modules/local/utility/resize_tif/main.nf

@@ -12,7 +12,7 @@ process RESIZE_TIF {
     output:
     tuple val(meta), path("${meta.id}/resized_*.tif"), emit: resized_mask
     tuple val("${task.process}"), val('python'), eval("python3 --version | sed 's/Python //'"), topic: versions, emit: versions_python
-    tuple val("${task.process}"), val('tifffile'), eval('python3 -c "import tifffile; print(tifffile.__version__)"'), topic: versions, emit: versions_tifffile
+    tuple val("${task.process}"), val('tifffile'), eval("pip show tifffile 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_tifffile
 
     when:
     task.ext.when == null || task.ext.when

modules/local/utility/upscale_mask/main.nf

@@ -27,7 +27,7 @@ process UPSCALE_MASK {
     output:
     tuple val(meta), path("${prefix}/upscaled_*.tif"), emit: upscaled_mask
     tuple val("${task.process}"), val('python'), eval("python3 --version | sed 's/Python //'"), topic: versions, emit: versions_python
-    tuple val("${task.process}"), val('tifffile'), eval('python3 -c "import tifffile; print(tifffile.__version__)"'), topic: versions, emit: versions_tifffile
+    tuple val("${task.process}"), val('tifffile'), eval("pip show tifffile 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_tifffile
 
     when:
     task.ext.when == null || task.ext.when

modules/local/xenium_patch/divide/main.nf

@@ -26,7 +26,7 @@ process XENIUM_PATCH_DIVIDE {
     tuple val(meta), path("patches/patch_grid.json")              , emit: grid
     tuple val(meta), path("patches/patch_*/transcripts.parquet")  , emit: patch_transcripts
     tuple val("${task.process}"), val('python'), eval("python3 --version | sed 's/Python //'"), topic: versions, emit: versions_python
-    tuple val("${task.process}"), val('pyarrow'), eval('python3 -c "import pyarrow; print(pyarrow.__version__)"'), topic: versions, emit: versions_pyarrow
+    tuple val("${task.process}"), val('pyarrow'), eval("pip show pyarrow 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_pyarrow
 
     when:
     task.ext.when == null || task.ext.when

modules/local/xenium_patch/stitch/main.nf

@@ -27,7 +27,7 @@ process XENIUM_PATCH_STITCH {
         path("output/xr-cell-polygons.geojson"),
         path("output/xr-transcript-metadata.csv")  , emit: xr_polygons_transcript
     tuple val("${task.process}"), val('python'), eval("python3 --version | sed 's/Python //'"), topic: versions, emit: versions_python
-    tuple val("${task.process}"), val('sopa'), eval('python3 -c "import sopa; print(sopa.__version__)"'), topic: versions, emit: versions_sopa
+    tuple val("${task.process}"), val('sopa'), eval("pip show sopa 2>/dev/null | sed -n 's/^Version: //p'"), topic: versions, emit: versions_sopa
 
     when:
     task.ext.when == null || task.ext.when