Code review for #52

.github/actions/nf-test/action.yml

@@ -38,11 +38,16 @@ runs:
     - name: Setup apptainer
       if: contains(inputs.profile, 'singularity')
       uses: eWaterCycle/setup-apptainer@4bb22c52d4f63406c49e94c804632975787312b3 # v2.0.0
+      with:
+        apptainer-version: 1.5.3
 
     - name: Set up Singularity
       if: contains(inputs.profile, 'singularity')
       shell: bash
       run: |
+        # The runner image restricts unprivileged user namespaces, which non-suid
+        # apptainer needs to start a container at all.
+        sudo sysctl -w kernel.apparmor_restrict_unprivileged_userns=0 || true
         mkdir -p $NXF_SINGULARITY_CACHEDIR
         mkdir -p $NXF_SINGULARITY_LIBRARYDIR
 

.github/workflows/awsfulltest.yml

@@ -24,7 +24,6 @@ jobs:
 
       - name: Launch workflow via Seqera Platform
         uses: seqeralabs/action-tower-launch@51565b514bff1827cf34620de25d0055759f1fc9 # v2
-        # TODO nf-core: You can customise AWS full pipeline tests as required
         # Add full size test data (but still relatively small datasets for few samples)
         # on the `test_full.config` test runs with only one set of parameters
         with:

.nf-core.yml

@@ -3,7 +3,9 @@ lint:
     - assets/nf-core-funcprofiler_logo_light.png
     - .github/PULL_REQUEST_TEMPLATE.md
     - .github/workflows/linting.yml
-    - .github/workflows/linting_comment.yml
+    # The 4.0.3 template renders `HEAD_REPO:` with two spaces, which the prettier
+    # pin used by prek collapses to one, so this file can never match the template.
+    - .github/workflows/branch.yml
     - .gitignore
 nf_core_version: 4.0.3
 repository_type: pipeline

assets/methods_description_template.yml

@@ -3,7 +3,6 @@ description: "Suggested text and references to use when describing pipeline usag
 section_name: "nf-core/funcprofiler Methods Description"
 section_href: "https://github.com/nf-core/funcprofiler"
 plot_type: "html"
-## TODO nf-core: Update the HTML below to your preferred methods description, e.g. add publication citation for this pipeline
 ## You inject any metadata in the Nextflow '${workflow}' object
 data: |
   <h4>Methods</h4>

conf/base.config

@@ -10,7 +10,6 @@
 
 process {
 
-    // TODO nf-core: Check the defaults for all processes
     cpus   = { 1      * task.attempt }
     memory = { 6.GB   * task.attempt }
     time   = { 4.h    * task.attempt }
@@ -24,7 +23,6 @@ process {
     //        These labels are used and recognised by default in DSL2 files hosted on nf-core/modules.
     //        If possible, it would be nice to keep the same label naming convention when
     //        adding in your local modules too.
-    // TODO nf-core: Customise requirements for specific processes.
     // See https://www.nextflow.io/docs/latest/config.html#config-process-selectors
     withLabel:process_single {
         cpus   = { 1                   }

conf/test_full.config

@@ -13,12 +13,8 @@
 params {
     config_profile_name                    = 'Full test profile'
     config_profile_description             = 'Full test dataset to check pipeline function'
-//    input                                  = params.pipelines_testdata_base_path + 'samplesheets/test.csv'
-//    databases                              = params.pipelines_testdata_base_path + 'database_full.csv'
     input                                  = 'https://raw.githubusercontent.com/nf-core/test-datasets/refs/heads/taxprofiler/samplesheet_shortreadsonly.csv'
     databases                              = 'https://raw.githubusercontent.com/nickp60/test-datasets/refs/heads/funcprofiler/database_full.csv'
-    outdir                                 = "fulltest_results"
-    cleanup                                = false
     run_fmhfunprofiler                     = true
     run_humann_v3                          = true
     run_mifaser                            = true

docs/CONTRIBUTING.md

@@ -182,4 +182,9 @@ If you update images or graphics, follow the nf-core [style guidelines](https://
 
 ## Pipeline specific contribution guidelines
 
-<!-- TODO nf-core: Add any pipeline specific contribution guidelines here, such as coding styles, procedures, checklists etc. -->
+When adding a new profiler:
+
+- Add it to [nf-core/modules](https://github.com/nf-core/modules) rather than to `modules/local`. A local module is only a stopgap for tools that are not yet packaged on Bioconda.
+- Gate it behind a `run_<tool>` parameter, and take its database from the database sheet so it is grouped by `dbprep` like every other tool.
+- Add it to the `profile` subworkflow, to `docs/output.md`, and to `CITATIONS.md`.
+- Cover it with an nf-test, and regenerate snapshots with `nf-test test --update-snapshot`.

main.nf

@@ -38,7 +38,7 @@ workflow NFCORE_FUNCPROFILER {
     //
     // WORKFLOW: Run pipeline
     //
-    FUNCPROFILER (
+    FUNCPROFILER(
         samplesheet,
         databases,
         params.multiqc_config,

modules/local/humann4/humann/main.nf

@@ -1,6 +1,6 @@
 // Taken 98% from https://github.com/nf-core/modules/pull/1089/files
 process HUMANN4 {
-    tag "$meta.id"
+    tag "${meta.id}"
     label 'process_high'
 
     conda "${moduleDir}/environment.yml"
@@ -14,10 +14,10 @@ process HUMANN4 {
     path utility_db
 
     output:
-    tuple val(meta), path("*_genefamilies.tsv.gz") , emit: genefamilies
+    tuple val(meta), path("*_genefamilies.tsv.gz"), emit: genefamilies
     tuple val(meta), path("*_pathabundance.tsv.gz"), emit: pathabundance
-    tuple val(meta), path("*_reactions.tsv.gz")    , emit: reactions
-    tuple val(meta), path("*.log")                 , emit: log
+    tuple val(meta), path("*_reactions.tsv.gz"), emit: reactions
+    tuple val(meta), path("*.log"), emit: log
     tuple val("${task.process}"), val('HUMAnN'), eval("humann --version 2>&1 | sed 's/humann v//'"), emit: versions_humann, topic: versions
     tuple val("${task.process}"), val('MetaPHLan'), eval("metaphlan --version 2>&1 | sed 's/metaphlan v//'"), emit: versions_metaphlan, topic: versions
 
@@ -39,13 +39,13 @@ process HUMANN4 {
 
     find \${NUCS_DB}
     humann \\
-        $args \\
+        ${args} \\
         --threads ${task.cpus} \\
-        --input $input \\
+        --input ${input} \\
         --protein-database \${PROTS_DB} \\
         --nucleotide-database \${NUCS_DB} \\
-        --output-basename $prefix \\
-        $pangenome_string \\
+        --output-basename ${prefix} \\
+        ${pangenome_string} \\
 	${args} \\
         --o-log ${prefix}.log \\
         --output .
@@ -54,11 +54,12 @@ process HUMANN4 {
     gzip -n *.tsv
 
     """
+
     stub:
     def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
-    echo $args
+    echo ${args}
 
     for suf in genefamilies.tsv.gz pathabundance.tsv.gz  reactions.tsv.gz
     do

nextflow_schema.json

@@ -129,19 +129,16 @@
                 },
                 "run_rgi": {
                     "type": "boolean",
-                    "description": "Turn on profiling with RGI. Requires database to be present CSV file passed to --databases",
-                    "default": false
+                    "description": "Turn on profiling with RGI. Requires database to be present CSV file passed to --databases"
                 },
                 "run_mifaser": {
                     "type": "boolean",
-                    "description": "Turn on profiling with mi-faser. Requires database to be present CSV file passed to --databases",
-                    "default": false
+                    "description": "Turn on profiling with mi-faser. Requires database to be present CSV file passed to --databases"
                 },
                 "run_eggnogmapper": {
                     "type": "boolean",
                     "fa_icon": "fas fa-toggle-on",
-                    "description": "Turn on functional annotation with EggNOG-mapper. Requires `eggnogmapper`-tagged database entries in the CSV file passed to --databases",
-                    "default": false
+                    "description": "Turn on functional annotation with EggNOG-mapper. Requires `eggnogmapper`-tagged database entries in the CSV file passed to --databases"
                 }
             }
         },

subworkflows/local/dataprep/meta.yml

@@ -1,50 +1,42 @@
 # yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/subworkflows/yaml-schema.json
 name: "dataprep"
-## TODO nf-core: Add a description of the subworkflow and list keywords
-description: Sort SAM/BAM/CRAM file
+description: |
+  Validate the input samplesheet and prepare read channels for profiling. Sets
+  `single_end`, `run_accession` and `instrument_platform` on the meta map, merges
+  multiple runs belonging to the same sample, and produces an additional channel
+  in which read pairs are concatenated into a single file for tools that cannot
+  take paired input.
 keywords:
-  - sort
-  - bam
-  - sam
-  - cram
-## TODO nf-core: Add a list of the modules and/or subworkflows used in the subworkflow
+  - samplesheet
+  - fastq
+  - merging
+  - concatenation
+  - metagenome
 components:
-  - samtools/sort
-  - samtools/index
-## TODO nf-core: List all of the channels used as input with a description and their structure
+  - cat/fastq
 input:
-  - ch_bam:
+  - samplesheet:
       type: file
       description: |
-        The input channel containing the BAM/CRAM/SAM files
-        Structure: [ val(meta), path(bam) ]
-      pattern: "*.{bam/cram/sam}"
-## TODO nf-core: List all of the channels used as output with a descriptions and their structure
+        Channel of validated samplesheet rows. Long-read (`OXFORD_NANOPORE`)
+        platforms and FASTA input are rejected, and `fastq_1` is required.
+        Structure: [ val(meta), val(run_accession), val(instrument_platform), path(fastq_1), path(fastq_2), path(fasta) ]
+      pattern: "*.{fq,fastq}{,.gz}"
 output:
-  - bam:
+  - reads:
       type: file
       description: |
-        Channel containing BAM files
-        Structure: [ val(meta), path(bam) ]
-      pattern: "*.bam"
-  - bai:
+        Channel of reads per sample, with runs merged where a sample has more
+        than one. Paired-end samples carry two files, single-end samples one.
+        Structure: [ val(meta), [ path(reads) ] ]
+      pattern: "*.{fq,fastq}{,.gz}"
+  - reads_concat:
       type: file
       description: |
-        Channel containing indexed BAM (BAI) files
-        Structure: [ val(meta), path(bai) ]
-      pattern: "*.bai"
-  - csi:
-      type: file
-      description: |
-        Channel containing CSI files
-        Structure: [ val(meta), path(csi) ]
-      pattern: "*.csi"
-  - versions:
-      type: file
-      description: |
-        File containing software versions
-        Structure: [ path(versions.yml) ]
-      pattern: "versions.yml"
+        Channel of the same reads concatenated into a single file per sample,
+        with `single_end` forced to `true`, for tools that accept only one FASTQ.
+        Structure: [ val(meta), [ path(reads) ] ]
+      pattern: "*.{fq,fastq}{,.gz}"
 authors:
   - "@nickp60"
 maintainers:

subworkflows/local/dbprep/meta.yml

@@ -1,50 +1,35 @@
 # yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/subworkflows/yaml-schema.json
 name: "dbprep"
-## TODO nf-core: Add a description of the subworkflow and list keywords
-description: Sort SAM/BAM/CRAM file
+description: |
+  Prepare the databases declared in the database sheet. Archives are decompressed
+  once per file and only for tools the user has enabled, then all entities
+  belonging to the same database are grouped so that each profiler receives a
+  single entry per database name and parameter combination.
 keywords:
-  - sort
-  - bam
-  - sam
-  - cram
-## TODO nf-core: Add a list of the modules and/or subworkflows used in the subworkflow
+  - database
+  - untar
+  - decompression
+  - metagenome
 components:
-  - samtools/sort
-  - samtools/index
-## TODO nf-core: List all of the channels used as input with a description and their structure
+  - untar
 input:
-  - ch_bam:
+  - databases:
       type: file
       description: |
-        The input channel containing the BAM/CRAM/SAM files
-        Structure: [ val(meta), path(bam) ]
-      pattern: "*.{bam/cram/sam}"
-## TODO nf-core: List all of the channels used as output with a descriptions and their structure
+        Channel of database entries from the database sheet. `meta.tool` selects
+        which profiler the database belongs to, `meta.db_entity` names the part
+        of a multi-file database, and `meta.db_params` carries tool-specific
+        arguments. Entries ending in `.tar`, `.tar.gz`, `.tar.bz2` or `.tar.xz`
+        are decompressed, the rest are passed through untouched.
+        Structure: [ val(meta), path(database) ]
+      pattern: "*.{tar,tar.gz,tar.bz2,tar.xz}"
 output:
-  - bam:
+  - dbs:
       type: file
       description: |
-        Channel containing BAM files
-        Structure: [ val(meta), path(bam) ]
-      pattern: "*.bam"
-  - bai:
-      type: file
-      description: |
-        Channel containing indexed BAM (BAI) files
-        Structure: [ val(meta), path(bai) ]
-      pattern: "*.bai"
-  - csi:
-      type: file
-      description: |
-        Channel containing CSI files
-        Structure: [ val(meta), path(csi) ]
-      pattern: "*.csi"
-  - versions:
-      type: file
-      description: |
-        File containing software versions
-        Structure: [ path(versions.yml) ]
-      pattern: "versions.yml"
+        Channel of databases grouped by tool, name and parameters. The second
+        element is a list of maps, each holding the entity name and its path.
+        Structure: [ val(meta), [ [ db_entity: val(entity), db_path: path(db) ] ] ]
 authors:
   - "@nickp60"
 maintainers: