Hificnv—copy-number variant caller for PacBio HiFi long-read data

modules/nf-core/hificnv/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:
+  - "bioconda::hificnv=1.0.1"

modules/nf-core/hificnv/main.nf

@@ -0,0 +1,68 @@
+process HIFICNV {
+    tag "$meta.id"
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'oras://community.wave.seqera.io/library/hificnv:1.0.1--ba0f622849fdd2bf':
+        'community.wave.seqera.io/library/hificnv:1.0.1--b7e433ac6789e2d2' }"
+
+    input:
+    tuple val(meta), path(bam), path(bai), path(maf)
+    tuple val(meta2), path(ref)
+    tuple val(meta3), path(exclude)
+    tuple val(meta4), path(expected_cn)
+
+    output:
+    tuple val(meta), path("*.copynum.bedgraph"), emit: copynum
+    tuple val(meta), path("*.depth.bw"),         emit: depth
+    tuple val(meta), path("*.maf.bw"),           emit: maf, optional: true
+    tuple val(meta), path("*.vcf.gz"),           emit: vcf
+    path "versions.yml",                         emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args   = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    // handle optional inputs
+    def maf_arg         = maf         ? "--maf ${maf}"                 : ""
+    def exclude_arg     = exclude     ? "--exclude ${exclude}"         : ""
+    def expected_cn_arg = expected_cn ? "--expected-cn ${expected_cn}" : ""
+
+    """
+    hificnv \\
+        --bam ${bam} \\
+        --ref ${ref} \\
+        ${maf_arg} \\
+        ${exclude_arg} \\
+        ${expected_cn_arg} \\
+        --threads ${task.cpus} \\
+        --output-prefix ${prefix} \\
+        ${args}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        hificnv: \$(hificnv --version 2>&1 | sed 's/^.*hificnv //; s/ .*\$//')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def create_maf = maf ? "touch ${prefix}.maf.bw" : ""
+
+    """
+    touch ${prefix}.depth.bw
+    touch ${prefix}.copynum.bedgraph
+    ${create_maf}
+    echo "" | gzip > ${prefix}.vcf.gz
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        hificnv: \$(hificnv --version 2>&1 | sed 's/^.*hificnv //; s/ .*\$//')
+    END_VERSIONS
+    """
+}

modules/nf-core/hificnv/meta.yml

@@ -0,0 +1,121 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "hificnv"
+description: "Copy number variant calling from PacBio HiFi reads"
+keywords:
+  - "copy number variation"
+  - "cnv"
+  - "PacBio"
+  - "HiFi"
+  - "long reads"
+  - "structural variation"
+tools:
+  - "hificnv":
+      description: "Copy number variant caller designed for PacBio HiFi reads"
+      homepage: "https://github.com/PacificBiosciences/HiFiCNV"
+      documentation: "https://github.com/PacificBiosciences/HiFiCNV"
+      tool_dev_url: "https://github.com/PacificBiosciences/HiFiCNV"
+      doi: "10.1093/bioinformatics/btac808"
+      licence: ["Pacific Biosciences Software License Agreement"]
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'test', single_end:false ]`
+    - bam:
+        type: file
+        description: "Sorted BAM or CRAM file from PacBio HiFi reads"
+        pattern: "*.{bam,cram}"
+    - bai:
+        type: file
+        description: BAM index file (CSI, CRAI, or BAI format)
+        pattern: "*.{bai,csi,crai}"
+    - maf:
+        type: file
+        description: "Minor allele frequency file (VCF format)"
+        pattern: "*.{vcf,vcf.gz}"
+        optional: true
+  - - meta2:
+        type: map
+        description: |
+          Groovy Map containing reference genome information
+          e.g. `[ id:'GATK.GRCh38' ]`
+    - ref:
+        type: file
+        description: "Reference genome FASTA file"
+        pattern: "*.{fa,fasta,fa.gz,fasta.gz}"
+  - - meta3:
+        type: map
+        description: |
+          Groovy Map containing exclude regions information
+          e.g. `[ id:'excluded_regions' ]`
+    - exclude:
+        type: file
+        description: "BED file containing regions to exclude from CNV calling"
+        pattern: "*.{bed,bed.gz}"
+        optional: true
+  - - meta4:
+        type: map
+        description: |
+          Groovy Map containing expected copy number information
+          e.g. `[ id:'male_expected_cn' ]`
+    - expected_cn:
+        type: file
+        description: "BED file containing expected copy number regions"
+        pattern: "*.{bed,bed.gz}"
+        optional: true
+
+output:
+  copynum:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'test', single_end:false ]`
+      - "*.copynum.bedgraph":
+          type: file
+          description: "Copy number bedGraph file"
+          pattern: "*.copynum.bedgraph"
+  depth:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'test', single_end:false ]`
+      - "*.depth.bw":
+          type: file
+          description: "Depth coverage bigWig file"
+          pattern: "*.depth.bw"
+  maf:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'test', single_end:false ]`
+      - "*.maf.bw":
+          type: file
+          description: "Minor allele frequency bigWig file"
+          pattern: "*.maf.bw"
+          optional: true
+  vcf:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'test', single_end:false ]`
+      - "*.vcf.gz":
+          type: file
+          description: "Copy number variants in VCF format"
+          pattern: "*.{vcf.gz}"
+  versions:
+    - "versions.yml":
+        type: file
+        description: "File containing software versions"
+        pattern: "versions.yml"
+
+authors:
+  - "@chaochaowong"
+
+maintainers:
+  - "@chaochaowong"