modules for new gbcms

.github/skip_nf_test.json

@@ -5,6 +5,11 @@
         "modules/msk/facets",
         "modules/msk/fgbio/collectduplexseqmetrics",
         "modules/msk/gbcms",
+        "modules/msk/gbcmsrs/dna",
+        "modules/msk/gbcmsrs/rna",
+        "modules/msk/gbcmsrs/normalize",
+        "modules/msk/gbcmsrs/merge",
+        "modules/msk/gbcmsrs/buildgtfcache",
         "modules/msk/neoantigenediting/computefitness",
         "modules/msk/neoantigenediting/aligntoiedb",
         "modules/msk/neoantigenutils/neoantigeninput",

modules/msk/gbcmsrs/buildgtfcache/environment.yml

@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "YOUR-TOOL=HERE"

modules/msk/gbcmsrs/buildgtfcache/main.nf

@@ -0,0 +1,48 @@
+process GBCMSRS_BUILDGTFCACHE {
+    tag "${variants.name}"
+    label 'process_single'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'ghcr.io/mskcc-omics-workflows/gbcms:6.3.1':
+        'ghcr.io/mskcc-omics-workflows/gbcms:6.3.1' }"
+
+    input:
+    path variants
+    path gtf
+
+    output:
+    path "gbcms_gtf_cache", emit: cache_dir
+    path "versions.yml"   , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error "GBCMSRS_BUILDGTFCACHE module does not support Conda. Please use Docker / Singularity instead."
+    }
+    def args = task.ext.args ?: ''
+    """
+    mkdir -p gbcms_gtf_cache
+    gbcms build-gtf-cache \\
+        --gtf ${gtf} \\
+        --variants ${variants} \\
+        --gtf-cache-dir gbcms_gtf_cache \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gbcms: \$(gbcms --version | sed 's/^gbcms //')
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    mkdir -p gbcms_gtf_cache
+    touch gbcms_gtf_cache/gbcms-gtf-stub.idx
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gbcms: \$(echo "${task.container}" | sed 's/.*://')
+    END_VERSIONS
+    """
+}

modules/msk/gbcmsrs/buildgtfcache/meta.yml

@@ -0,0 +1,54 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/yaml-schema.json
+name: "gbcmsrs_buildgtfcache"
+description:
+  Pre-build the gbcms GTF index cache once for a cohort so that per-sample
+  `gbcms rna` runs skip re-parsing the GTF
+keywords:
+  - cache
+  - gtf
+  - rna
+  - index
+tools:
+  - "gbcms":
+      description:
+        "A high-performance, orientation-aware genotype counting system for
+        genomic variants (Rust rewrite of GetBaseCountsMultiSample)"
+      homepage: "https://github.com/msk-access/gbcms"
+      documentation: "https://msk-access.github.io/gbcms/"
+      tool_dev_url: "https://github.com/msk-access/gbcms"
+      licence: ["AGPL-3.0"]
+      identifier: ""
+
+input:
+  - variants:
+      type: file
+      description:
+        Variant file (VCF/MAF) for the cohort. Only its chromosome set is
+        used, and it must be the same variant file the per-sample `gbcms rna` runs use
+        so the cache key lines up.
+      pattern: "*.{vcf,maf}"
+      ontologies: []
+  - gtf:
+      type: file
+      description: GTF annotation file (Ensembl/GENCODE)
+      pattern: "*.gtf"
+      ontologies: []
+output:
+  cache_dir:
+    - gbcms_gtf_cache:
+        type: directory
+        description:
+          Directory containing the serialized GTF index cache. Point every
+          per-sample `gbcms rna --gtf-cache-dir` at this same directory.
+        pattern: "gbcms_gtf_cache"
+  versions:
+    - versions.yml:
+        type: file
+        description: File containing software versions
+        pattern: "versions.yml"
+        ontologies:
+          - edam: http://edamontology.org/format_3750 # YAML
+authors:
+  - "@shahr"
+maintainers:
+  - "@shahr"

modules/msk/gbcmsrs/buildgtfcache/tests/main.nf.test

@@ -0,0 +1,35 @@
+// nf-core modules test gbcmsrs/buildgtfcache
+nextflow_process {
+
+    name "Test Process GBCMSRS_BUILDGTFCACHE"
+    script "../main.nf"
+    process "GBCMSRS_BUILDGTFCACHE"
+
+    tag "modules"
+    tag "modules_msk"
+    tag "gbcmsrs"
+    tag "gbcmsrs/buildgtfcache"
+
+    test("sarscov2 illumina - vcf gtf") {
+
+        when {
+            process {
+                """
+                input[0] = file(params.test_data['sarscov2']['illumina']['test_vcf'], checkIfExists: true)
+                input[1] = file(params.test_data['sarscov2']['genome']['genome_gtf'], checkIfExists: true)
+                """
+            }
+        }
+
+        then {
+            def idx_files = file(process.out.cache_dir[0]).listFiles().findAll { it.name.endsWith('.idx') }
+            assertAll(
+                { assert process.success },
+                { assert idx_files.size() == 1 },
+                { assert snapshot(process.out.versions).match() }
+            )
+        }
+
+    }
+
+}

modules/msk/gbcmsrs/buildgtfcache/tests/main.nf.test.snap

@@ -0,0 +1,14 @@
+{
+    "sarscov2 illumina - vcf gtf": {
+        "content": [
+            [
+                "versions.yml:md5,9d75c19a947df6e008a3918a2e97ebc6"
+            ]
+        ],
+        "timestamp": "2026-08-27T10:14:42.910180254",
+        "meta": {
+            "nf-test": "0.9.5",
+            "nextflow": "26.04.6"
+        }
+    }
+}

modules/msk/gbcmsrs/buildgtfcache/tests/tags.yml

@@ -0,0 +1,2 @@
+gbcmsrs/buildgtfcache:
+  - "modules/msk/gbcmsrs/buildgtfcache/**"

modules/msk/gbcmsrs/dna/environment.yml

@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "YOUR-TOOL=HERE"

modules/msk/gbcmsrs/dna/main.nf

@@ -0,0 +1,61 @@
+process GBCMSRS_DNA {
+    tag "$meta.id"
+    label 'process_medium'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'ghcr.io/mskcc-omics-workflows/gbcms:6.3.1':
+        'ghcr.io/mskcc-omics-workflows/gbcms:6.3.1' }"
+
+    input:
+    tuple val(meta), path(variants), val(sample_names), path(bams), path(bais)
+    path fasta
+    path fasta_fai
+
+    output:
+    tuple val(meta), path("gbcms_out/*.{vcf,maf}"), emit: variant_file
+    path "versions.yml"                            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error "GBCMSRS_DNA module does not support Conda. Please use Docker / Singularity instead."
+    }
+    // gbcms's read filters (--filter-duplicates/-secondary/-supplementary/-qc-failed)
+    // default to ON. A pipeline mapping its own boolean params into ext.args must emit
+    // the explicit --no-filter-x form when off; omitting the flag silently keeps it on.
+    // On Nextflow >=26.04 (strict parser), CLI param overrides arrive as Strings, so
+    // `params.x ? 'a' : 'b'` sees "false" as truthy — compare with `.toString() == 'true'`.
+    def args = task.ext.args ?: ''
+    // Bare `--bam path` labels the sample using the staged file's stem, which is not
+    // meaningful for real BAM naming conventions. Pairing each bam with an explicit
+    // name keeps the output filename and Tumor_Sample_Barcode/VCF sample column
+    // predictable and equal to what the caller intends.
+    def bam_args = [sample_names, bams].transpose().collect { name, bam -> "--bam ${name}:${bam}" }.join(' ')
+    """
+    gbcms dna \\
+        --variants ${variants} \\
+        ${bam_args} \\
+        --fasta ${fasta} \\
+        --output-dir gbcms_out \\
+        --threads ${task.cpus} \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gbcms: \$(gbcms --version | sed 's/^gbcms //')
+    END_VERSIONS
+    """
+
+    stub:
+    def sample_name = sample_names[0]
+    """
+    mkdir -p gbcms_out
+    touch gbcms_out/${sample_name}.vcf
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gbcms: \$(echo "${task.container}" | sed 's/.*://')
+    END_VERSIONS
+    """
+}

modules/msk/gbcmsrs/dna/meta.yml

@@ -0,0 +1,84 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/yaml-schema.json
+name: "gbcmsrs_dna"
+description:
+  Count alleles in cfDNA/somatic DNA BAMs at known variant sites using
+  gbcms
+keywords:
+  - basecount
+  - genotyping
+  - dna
+  - variants
+tools:
+  - "gbcms":
+      description:
+        "A high-performance, orientation-aware genotype counting system for
+        genomic variants (Rust rewrite of GetBaseCountsMultiSample)"
+      homepage: "https://github.com/msk-access/gbcms"
+      documentation: "https://msk-access.github.io/gbcms/"
+      tool_dev_url: "https://github.com/msk-access/gbcms"
+      licence: ["AGPL-3.0"]
+      identifier: ""
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'test' ]`
+    - variants:
+        type: file
+        description: Input variant file in VCF or MAF format
+        pattern: "*.{vcf,maf}"
+        ontologies: []
+    - sample_names:
+        type: string
+        description:
+          List of sample names, parallel to `bams`, used to label each BAM
+          via `--bam <name>:<path>`. Without an explicit name, gbcms falls
+          back to the staged file's stem, which is usually not the desired
+          sample id.
+    - bams:
+        type: file
+        description: One or more indexed BAM/CRAM files to count alleles in
+        pattern: "*.{bam,cram}"
+        ontologies: []
+    - bais:
+        type: file
+        description: Index file(s) for the input BAM/CRAM file(s)
+        pattern: "*.{bai,crai}"
+        ontologies: []
+  - fasta:
+      type: file
+      description: Reference genome FASTA file
+      pattern: "*.{fasta,fa}"
+      ontologies: []
+  - fasta_fai:
+      type: file
+      description: Index of the reference genome FASTA file
+      pattern: "*.fai"
+      ontologies: []
+output:
+  variant_file:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'test' ]`
+      - "gbcms_out/*.{vcf,maf}":
+          type: file
+          description:
+            Variant file annotated with allele counts, strand information,
+            and statistical annotations
+          pattern: "gbcms_out/*.{vcf,maf}"
+  versions:
+    - versions.yml:
+        type: file
+        description: File containing software versions
+        pattern: "versions.yml"
+        ontologies:
+          - edam: http://edamontology.org/format_3750 # YAML
+authors:
+  - "@shahr"
+maintainers:
+  - "@buehlere"
+  - "@shahr"

modules/msk/gbcmsrs/dna/tests/main.nf.test

@@ -0,0 +1,40 @@
+// nf-core modules test gbcmsrs/dna
+nextflow_process {
+
+    name "Test Process GBCMSRS_DNA"
+    script "../main.nf"
+    process "GBCMSRS_DNA"
+
+    tag "modules"
+    tag "modules_msk"
+    tag "gbcmsrs"
+    tag "gbcmsrs/dna"
+
+    test("sarscov2 illumina - vcf") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test' ], // meta map
+                    file(params.test_data['sarscov2']['illumina']['test_vcf'], checkIfExists: true),
+                    [ 'test' ],
+                    [ file(params.test_data['sarscov2']['illumina']['test_single_end_sorted_bam'], checkIfExists: true) ],
+                    [ file(params.test_data['sarscov2']['illumina']['test_single_end_sorted_bam_bai'], checkIfExists: true) ]
+                ]
+                input[1] = file(params.test_data['sarscov2']['genome']['genome_fasta'], checkIfExists: true)
+                input[2] = file(params.test_data['sarscov2']['genome']['genome_fasta_fai'], checkIfExists: true)
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}