Merge pull request #1724 from nvnieuwk/new-module-gatk4/CalibrateDragstrModel

new module gatk4/calibratedragstrmodel

modules/gatk4/calibratedragstrmodel/main.nf

@@ -0,0 +1,48 @@
+process GATK4_CALIBRATEDRAGSTRMODEL {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.6.1" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gatk4:4.2.6.1--hdfd78af_0':
+        'quay.io/biocontainers/gatk4:4.2.6.1--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(bam), path(bam_index)
+    path  fasta
+    path  fasta_fai
+    path  dict
+    path  strtablefile
+
+    output:
+    tuple val(meta), path("*.txt")   , emit: dragstr_model
+    path "versions.yml"              , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    def avail_mem = 3
+    if (!task.memory) {
+        log.info '[GATK CalibrateDragstrModel] Available memory not known - defaulting to 3GB. Specify process memory requirements to change this.'
+    } else {
+        avail_mem = task.memory.giga
+    }
+    """
+    gatk --java-options "-Xmx${avail_mem}g" CalibrateDragstrModel \\
+        --input $bam \\
+        --output ${prefix}.txt \\
+        --reference $fasta \\
+        --str-table-path $strtablefile \\
+        --tmp-dir . \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gatk4: \$(echo \$(gatk --version 2>&1) | sed 's/^.*(GATK) v//; s/ .*\$//')
+    END_VERSIONS
+    """
+}

modules/gatk4/calibratedragstrmodel/meta.yml

@@ -0,0 +1,70 @@
+name: gatk4_calibratedragstrmodel
+description: estimates the parameters for the DRAGstr model
+keywords:
+  - gatk4
+  - bam
+  - cram
+  - sam
+  - calibratedragstrmodel
+tools:
+  - gatk4:
+      description:
+        Genome Analysis Toolkit (GATK4). 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/360057441571-CalibrateDragstrModel-BETA-
+      tool_dev_url: https://github.com/broadinstitute/gatk
+      doi: 10.1158/1538-7445.AM2017-3590
+      licence: ["Apache-2.0"]
+
+input:
+  # Only when we have meta
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - bam:
+      type: file
+      description: BAM/CRAM/SAM file
+      pattern: "*.{bam,cram,sam}"
+  - bam_index:
+      type: file
+      description: index of the BAM/CRAM/SAM file
+      pattern: "*.{bai,crai,sai}"
+  - fasta:
+      type: file
+      description: The reference FASTA file
+      pattern: "*.{fasta,fa}"
+  - fasta_fai:
+      type: file
+      description: The index of the reference FASTA file
+      pattern: "*.fai"
+  - dict:
+      type: file
+      description: The sequence dictionary of the reference FASTA file
+      pattern: "*.dict"
+  - strtablefile:
+      type: file
+      description: The StrTableFile zip folder of the reference FASTA file
+      pattern: "*.zip"
+
+output:
+  #Only when we have meta
+  - 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"
+  - dragstr_model:
+      type: file
+      description: The DragSTR model
+      pattern: "*.txt"
+
+authors:
+  - "@nvnieuwk"

tests/config/pytest_modules.yml

@@ -743,6 +743,10 @@ gatk4/calculatecontamination:
   - modules/gatk4/calculatecontamination/**
   - tests/modules/gatk4/calculatecontamination/**
 
+gatk4/calibratedragstrmodel:
+  - modules/gatk4/calibratedragstrmodel/**
+  - tests/modules/gatk4/calibratedragstrmodel/**
+
 gatk4/cnnscorevariants:
   - modules/gatk4/cnnscorevariants/**
   - tests/modules/gatk4/cnnscorevariants/**

tests/config/test_data.config

@@ -123,6 +123,7 @@ params {
                 genome_elfasta                                 = "${test_data_dir}/genomics/homo_sapiens/genome/genome.elfasta"
                 genome_fasta                                   = "${test_data_dir}/genomics/homo_sapiens/genome/genome.fasta"
                 genome_fasta_fai                               = "${test_data_dir}/genomics/homo_sapiens/genome/genome.fasta.fai"
+                genome_strtablefile                            = "${test_data_dir}/genomics/homo_sapiens/genome/genome_strtablefile.zip"
                 genome_dict                                    = "${test_data_dir}/genomics/homo_sapiens/genome/genome.dict"
                 genome_gff3                                    = "${test_data_dir}/genomics/homo_sapiens/genome/genome.gff3"
                 genome_gtf                                     = "${test_data_dir}/genomics/homo_sapiens/genome/genome.gtf"

tests/modules/gatk4/calibratedragstrmodel/main.nf

@@ -0,0 +1,44 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { GATK4_CALIBRATEDRAGSTRMODEL } from '../../../../modules/gatk4/calibratedragstrmodel/main.nf'
+
+workflow test_gatk4_calibratedragstrmodel_bam {
+    
+    input = [
+        [ id:'test', single_end:false ], // meta map
+        file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_bam'], checkIfExists: true),
+        file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_bam_bai'], checkIfExists: true)
+    ]
+
+    fasta = file(params.test_data['homo_sapiens']['genome']['genome_fasta'], checkIfExists: true)
+
+    fasta_fai = file(params.test_data['homo_sapiens']['genome']['genome_fasta_fai'], checkIfExists: true)
+
+    dict = file(params.test_data['homo_sapiens']['genome']['genome_dict'], checkIfExists: true)
+
+    strtablefile = file(params.test_data['homo_sapiens']['genome']['genome_strtablefile'], checkIfExists: true)
+
+    GATK4_CALIBRATEDRAGSTRMODEL ( input, fasta, fasta_fai, dict, strtablefile )
+}
+
+workflow test_gatk4_calibratedragstrmodel_cram {
+    
+    input = [
+        [ id:'test', single_end:false ], // meta map
+        file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_cram'], checkIfExists: true),
+        file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_cram_crai'], checkIfExists: true)
+    ]
+
+    fasta = file(params.test_data['homo_sapiens']['genome']['genome_fasta'], checkIfExists: true)
+
+    fasta_fai = file(params.test_data['homo_sapiens']['genome']['genome_fasta_fai'], checkIfExists: true)
+
+    dict = file(params.test_data['homo_sapiens']['genome']['genome_dict'], checkIfExists: true)
+
+    strtablefile = file(params.test_data['homo_sapiens']['genome']['genome_strtablefile'], checkIfExists: true)
+
+    GATK4_CALIBRATEDRAGSTRMODEL ( input, fasta, fasta_fai, dict, strtablefile )
+}
+

tests/modules/gatk4/calibratedragstrmodel/nextflow.config

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

tests/modules/gatk4/calibratedragstrmodel/test.yml

@@ -0,0 +1,17 @@
+- name: gatk4 calibratedragstrmodel test_gatk4_calibratedragstrmodel_bam
+  command: nextflow run ./tests/modules/gatk4/calibratedragstrmodel -entry test_gatk4_calibratedragstrmodel_bam -c ./tests/config/nextflow.config  -c ./tests/modules/gatk4/calibratedragstrmodel/nextflow.config
+  tags:
+    - gatk4
+    - gatk4/calibratedragstrmodel
+  files:
+    - path: output/gatk4/test.txt
+      md5sum: 0a1a1583b157fa2251dd931ed165da4f
+
+- name: gatk4 calibratedragstrmodel test_gatk4_calibratedragstrmodel_cram
+  command: nextflow run ./tests/modules/gatk4/calibratedragstrmodel -entry test_gatk4_calibratedragstrmodel_cram -c ./tests/config/nextflow.config  -c ./tests/modules/gatk4/calibratedragstrmodel/nextflow.config
+  tags:
+    - gatk4
+    - gatk4/calibratedragstrmodel
+  files:
+    - path: output/gatk4/test.txt
+      md5sum: 1aa7ab38023f724877b3323c5e6b9a4e