Gatk4 selectvariants (#1346)

* initial commit

* tested

Co-authored-by: Cipriano <rrn8@cdc.gov>
Co-authored-by: Sateesh <33637490+sateeshperi@users.noreply.github.com>
Co-authored-by: Maxime U. Garcia <max.u.garcia@gmail.com>

modules/gatk4/selectvariants/main.nf

@@ -0,0 +1,41 @@
+process GATK4_SELECTVARIANTS {
+    tag "$meta.id"
+    label 'process_medium'
+
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gatk4:4.2.5.0--hdfd78af_0':
+        'quay.io/biocontainers/gatk4:4.2.5.0--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(vcf), path(vcf_idx)
+
+    output:
+    tuple val(meta), path("*.selectvariants.vcf.gz")       , emit: vcf
+    tuple val(meta), path("*.selectvariants.vcf.gz.tbi")   , emit: tbi
+    path "versions.yml"		                               , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def avail_mem = 3
+    if (!task.memory) {
+        log.info '[GATK VariantFiltration] Available memory not known - defaulting to 3GB. Specify process memory requirements to change this.'
+    } else {
+        avail_mem = task.memory.toGiga()
+    }
+    """
+    gatk --java-options "-Xmx${avail_mem}G" SelectVariants \\
+        -V $vcf \\
+        -O ${prefix}.selectvariants.vcf.gz \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gatk4: \$(echo \$(gatk --version 2>&1) | sed 's/^.*(GATK) v//; s/ .*\$//')
+    END_VERSIONS
+    """
+}

modules/gatk4/selectvariants/meta.yml

@@ -0,0 +1,55 @@
+name: gatk4_selectvariants
+description: Select a subset of variants from a VCF file
+keywords:
+  - gatk
+  - gatk4
+  - selectvariants
+  - vcf
+tools:
+  - gatk4:
+      description: |
+        Developed in the Data Sciences Platform at the Broad Institute, the toolkit offers a wide variety of tools
+        with a primary focus on variant discovery and genotyping. Its powerful processing engine
+        and high-performance computing features make it capable of taking on projects of any size.
+      homepage: https://gatk.broadinstitute.org/hc/en-us
+      documentation: https://gatk.broadinstitute.org/hc/en-us/articles/360036362532-SelectVariants
+      tool_dev_url: https://github.com/broadinstitute/gatk
+      doi: 10.1158/1538-7445.AM2017-3590
+      licence: ["Apache-2.0"]
+
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test']
+  - vcf:
+      type: list
+      description: VCF(.gz) file
+      pattern: "*.{vcf,vcf.gz}"
+  - vcf_idx:
+      type: list
+      description: VCF file index
+      pattern: "*.{idx,tbi}"
+
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - vcf:
+      type: file
+      description: Compressed VCF file
+      pattern: "*.selectvariants.vcf.gz"
+  - vcf_tbi:
+      type: list
+      description: VCF file index
+      pattern: "*.{idx,tbi}"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+
+authors:
+  - "@mjcipriano"

tests/config/pytest_modules.yml

@@ -672,6 +672,10 @@ gatk4/samtofastq:
   - modules/gatk4/samtofastq/**
   - tests/modules/gatk4/samtofastq/**
 
+gatk4/selectvariants:
+  - modules/gatk4/selectvariants/**
+  - tests/modules/gatk4/selectvariants/**
+
 gatk4/splitncigarreads:
   - modules/gatk4/splitncigarreads/**
   - tests/modules/gatk4/splitncigarreads/**

tests/modules/gatk4/selectvariants/main.nf

@@ -0,0 +1,29 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { GATK4_SELECTVARIANTS } from '../../../../modules/gatk4/selectvariants/main.nf'
+
+// Basic parameters with uncompressed VCF input
+workflow test_gatk4_selectvariants_vcf_input {
+
+    input = [
+        [ id:'test' ], // meta map
+        file(params.test_data['homo_sapiens']['illumina']['test_genome_vcf'], checkIfExists: true),
+        file(params.test_data['homo_sapiens']['illumina']['test_genome_vcf_idx'], checkIfExists: true)
+    ]
+
+    GATK4_SELECTVARIANTS ( input)
+}
+
+// Basic parameters with compressed VCF input
+workflow test_gatk4_selectvariants_gz_input {
+
+    input = [
+        [ id:'test' ], // meta map
+        file(params.test_data['homo_sapiens']['illumina']['test_genome_vcf_gz'], checkIfExists: true),
+        file(params.test_data['homo_sapiens']['illumina']['test_genome_vcf_gz_tbi'], checkIfExists: true)
+    ]
+
+    GATK4_SELECTVARIANTS ( input )
+}

tests/modules/gatk4/selectvariants/nextflow.config

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

tests/modules/gatk4/selectvariants/test.yml

@@ -0,0 +1,21 @@
+- name: gatk4 selectvariants test_gatk4_selectvariants_vcf_input
+  command: nextflow run tests/modules/gatk4/selectvariants -entry test_gatk4_selectvariants_vcf_input -c tests/config/nextflow.config
+  tags:
+    - gatk4/selectvariants
+    - gatk4
+  files:
+    - path: output/gatk4/test.selectvariants.vcf.gz
+    - path: output/gatk4/test.selectvariants.vcf.gz.tbi
+    - path: output/gatk4/versions.yml
+      md5sum: a35d78af179f43652274bc7405d5a785
+
+- name: gatk4 selectvariants test_gatk4_selectvariants_gz_input
+  command: nextflow run tests/modules/gatk4/selectvariants -entry test_gatk4_selectvariants_gz_input -c tests/config/nextflow.config
+  tags:
+    - gatk4/selectvariants
+    - gatk4
+  files:
+    - path: output/gatk4/test.selectvariants.vcf.gz
+    - path: output/gatk4/test.selectvariants.vcf.gz.tbi
+    - path: output/gatk4/versions.yml
+      md5sum: c943f3579a369968ca63444eb43fb6e7