Picard cleansam (#1306)

* add picard-cleansam

* add picard/cleansam

* update test yml with output

* picard 2.26.10 -> 2.26.9

* add output to test yml

Co-authored-by: Peri <rrx8@cdc.gov>
Co-authored-by: Robert A. Petit III <robbie.petit@gmail.com>

modules/picard/cleansam/main.nf

@@ -0,0 +1,44 @@
+process PICARD_CLEANSAM {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda (params.enable_conda ? "bioconda::picard=2.26.9" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/picard:2.26.9--hdfd78af_0' :
+        'quay.io/biocontainers/picard:2.26.9--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(sam)
+
+    output:
+    tuple val(meta), path("*.sam"), emit: sam
+    path "versions.yml"           , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def STRINGENCY = task.ext.stringency ?: "STRICT"
+    def avail_mem = 3
+    if (!task.memory) {
+        log.info '[Picard CleanSam] Available memory not known - defaulting to 3GB. Specify process memory requirements to change this.'
+    } else {
+        avail_mem = task.memory.giga
+    }
+    """
+    picard \\
+        -Xmx${avail_mem}g \\
+        CleanSam  \\
+        ${args} \\
+        -I ${sam} \\
+        -O ${prefix}.sam \\
+        --VALIDATION_STRINGENCY ${STRINGENCY}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        picard: \$(picard CleanSam --version 2>&1 | grep -o 'Version:.*' | cut -f2- -d:)
+    END_VERSIONS
+    """
+}

modules/picard/cleansam/meta.yml

@@ -0,0 +1,46 @@
+name: picard_cleansam
+description: Cleans the provided SAM/BAM, soft-clipping beyond-end-of-reference alignments and setting MAPQ to 0 for unmapped reads
+keywords:
+  - clean
+  - sam
+  - bam
+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/360036491452-CleanSam-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 ]
+  - sam:
+      type: file
+      description: SAM file
+      pattern: "*.{sam}"
+
+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"
+  - sam:
+      type: file
+      description: Cleaned SAM file
+      pattern: "*.{sam}"
+
+authors:
+  - "@sateeshperi"
+  - "@mjcipriano"
+  - "@hseabolt"

tests/config/pytest_modules.yml

@@ -1157,6 +1157,10 @@ phyloflash:
   - modules/phyloflash/**
   - tests/modules/phyloflash/**
 
+picard/cleansam:
+  - modules/picard/cleansam/**
+  - tests/modules/picard/cleansam/**
+
 picard/collecthsmetrics:
   - modules/picard/collecthsmetrics/**
   - tests/modules/picard/collecthsmetrics/**

tests/modules/picard/cleansam/main.nf

@@ -0,0 +1,15 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { PICARD_CLEANSAM } from '../../../../modules/picard/cleansam/main.nf'
+
+workflow test_picard_cleansam {
+    
+    input = [ 
+        [ id:'test', single_end:true ], // meta map
+        file(params.test_data['sarscov2']['illumina']['test_single_end_bam'], checkIfExists: true) 
+    ]
+
+    PICARD_CLEANSAM ( input )
+}

tests/modules/picard/cleansam/nextflow.config

@@ -0,0 +1,5 @@
+process {
+
+    publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
+    
+}

tests/modules/picard/cleansam/test.yml

@@ -0,0 +1,10 @@
+- name: picard cleansam test_picard_cleansam
+  command: nextflow run tests/modules/picard/cleansam -entry test_picard_cleansam -c tests/config/nextflow.config
+  tags:
+    - picard/cleansam
+    - picard
+  files:
+    - path: output/picard/test.sam
+      md5sum: e314171a6060eb79947c13ad126ddf00
+    - path: output/picard/versions.yml
+      md5sum: e6457d7c6de51bf6f4b577eda65e57ac