Merge pull request #21 from CCBR/picard-samtofastq

New module picard samtofastq

CHANGELOG.md

@@ -8,4 +8,5 @@
 - custom/bam_to_fastq (#14)
 - cutadapt (#11)
 - khmer/uniquekmers (#7)
+- picard/samtofastq (#21)
 - samtools/filteraligned (#13,#20)

README.md

@@ -38,5 +38,6 @@ Many of the modules and subworkflows in this project reuse and adapt code from [
 In those cases, credit is noted in the `meta.yml` file of the module/subworkflow and also listed here:
 
 - [bwa](modules/CCBR/bwa) adapts the [nf-core bwa module](https://github.com/nf-core/chipseq/tree/51eba00b32885c4d0bec60db3cb0a45eb61e34c5/modules/nf-core/modules/bwa)
-- [cutadapt](modules/CCBR/cutadapt) adapts the [nf-core cutadapt module](https://github.com/nf-core/modules/tree/master/modules/nf-core/cutadapt)
+- [cutadapt](modules/CCBR/cutadapt) adapts the [nf-core cutadapt module](https://github.com/nf-core/modules/tree/ef007b1ce5316506b8c27c3e7a62482409c6153c/modules/nf-core/cutadapt)
 - [khmer](modules/CCBR/khmer) adapts the [nf-core khmer module](https://github.com/nf-core/modules/tree/b48a1efc8e067502e1a9bafbac788c1e0abdfc6a/modules/nf-core/khmer)
+- [picard/samtofastq](modules/picard/samtofastq) adapts the [nf-core gatk4 samtofastq module](https://github.com/nf-core/modules/tree/ef007b1ce5316506b8c27c3e7a62482409c6153c/modules/nf-core/gatk4/samtofastq)

modules/CCBR/picard/samtofastq/main.nf

@@ -0,0 +1,48 @@
+process PICARD_SAMTOFASTQ {
+    tag { meta.id }
+    label 'process_medium'
+
+    container 'nciccbr/ccbr_picard_2.27.5:v1'
+
+    input:
+    tuple val(meta), path(bam)
+
+    output:
+    tuple val(meta), path("*.fastq.gz"), emit: reads
+    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}"
+    def output = meta.single_end ? "--FASTQ ${prefix}.fastq" : "--FASTQ ${prefix}_1.fastq --SECOND_END_FASTQ ${prefix}_2.fastq --UNPAIRED_FASTQ ${prefix}.unpaired.fastq"
+
+    if (!task.memory) {
+        log.warn '[Picard SamToFastq] Available memory not known - defaulting to 3GB. Specify process memory requirements to change this.'
+    }
+    def avail_mem = task.memory ? (task.memory.mega*0.8).intValue() : 3072
+
+    """
+    picard \\
+        -Xmx${avail_mem}M \\
+        SamToFastq \\
+        ${args} \\
+        --INPUT ${bam} \\
+        ${output}
+
+    pigz -p ${task.cpus} *.fastq
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        picard: \$(picard SamToFastq --version 2>&1 | grep -o 'Version:.*' | cut -f2- -d:)
+    END_VERSIONS
+    """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.fastq.gz versions.yml
+    """
+}

modules/CCBR/picard/samtofastq/meta.yml

@@ -0,0 +1,44 @@
+name: "picard_samtofastq"
+description: Converts a BAM or SAM file to fastq. Adapted from the nf-core gatk4 samtofastq module.
+keywords:
+  - fastq
+  - bam
+  - sam
+tools:
+  - picard:
+      description: |
+        A set of command line tools (in Java) for manipulating high-throughput sequencing (HTS)
+        data and formats such as SAM/BAM/CRAM and VCF.
+      homepage: https://broadinstitute.github.io/picard/
+      documentation: https://gatk.broadinstitute.org/hc/en-us/articles/360036510672-FastqToSam-Picard-
+      tool_dev_url: https://github.com/broadinstitute/picard
+      licence: ["MIT"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - bam:
+      type: file
+      description: |
+        SAM or BAM file
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  #
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+  - fastq:
+      type: file
+      description: fastq files
+      pattern: "*.{fastq}"
+authors:
+  - "@kelly-sovacool"
+maintainers:
+  - "@kelly-sovacool"

tests/config/pytest_modules.yml

@@ -18,6 +18,10 @@ khmer/uniquekmers:
   - modules/CCBR/khmer/uniquekmers/**
   - tests/CCBR/khmer/uniquekmers/**
 
+picard/samtofastq:
+  - modules/CCBR/picard/samtofastq/**
+  - tests/modules/CCBR/picard/samtofastq/**
+
 samtools/filteraligned:
   - modules/CCBR/samtools/filteraligned/**
   - tests/modules/CCBR/samtools/filteraligned/**

tests/modules/CCBR/picard/samtofastq/main.nf

@@ -0,0 +1,19 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { PICARD_SAMTOFASTQ } from '../../../../../modules/CCBR/picard/samtofastq/main.nf'
+
+workflow test_picard_samtofastq_single {
+    input = [ [ id:'test', single_end: true ], // meta map
+              [ file(params.test_data['sarscov2']['illumina']['test_single_end_bam'], checkIfExists: true) ]
+            ]
+    PICARD_SAMTOFASTQ ( input )
+}
+
+workflow test_picard_samtofastq_paired {
+    input = [ [ id:'test', single_end: false ], // meta map
+              [ file(params.test_data['sarscov2']['illumina']['test_paired_end_bam'], checkIfExists: true) ]
+            ]
+    PICARD_SAMTOFASTQ ( input )
+}

tests/modules/CCBR/picard/samtofastq/nextflow.config

@@ -0,0 +1,4 @@
+process {
+
+    publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
+}

tests/modules/CCBR/picard/samtofastq/test.yml

@@ -0,0 +1,34 @@
+- name: picard samtofastq test_picard_samtofastq_single
+  command: nextflow run ./tests/modules/CCBR/picard/samtofastq -entry test_picard_samtofastq_single -c ./tests/config/nextflow.config
+  tags:
+    - picard/samtofastq
+    - picard
+  files:
+    - path: output/picard/test.fastq.gz
+      contains:
+        - "@ERR5069949.2151832"
+    - path: output/picard/versions.yml
+
+- name: picard samtofastq test_picard_samtofastq_paired
+  command: nextflow run ./tests/modules/CCBR/picard/samtofastq -entry test_picard_samtofastq_paired -c ./tests/config/nextflow.config
+  tags:
+    - picard/samtofastq
+    - picard
+  files:
+    - path: output/picard/test_1.fastq.gz
+      contains:
+        - "@ERR5069949.2151832/1"
+    - path: output/picard/test_2.fastq.gz
+      contains:
+        - "@ERR5069949.2151832/2"
+    - path: output/picard/test.unpaired.fastq.gz
+    - path: output/picard/versions.yml
+
+- name: picard samtofastq test_picard_samtofastq_single stub
+  command: nextflow run ./tests/modules/CCBR/picard/samtofastq -entry test_picard_samtofastq_single -c ./tests/config/nextflow.config -stub
+  tags:
+    - picard/samtofastq
+    - picard
+  files:
+    - path: output/picard/test.fastq.gz
+    - path: output/picard/versions.yml