implementation of goleft/indexcov

conf/modules/goleft.indexcov.config

@@ -0,0 +1,26 @@
+/*
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Config file for defining DSL2 per module options and publishing paths
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Available keys to override module options:
+        ext.args   = Additional arguments appended to command in module.
+        ext.prefix = File name prefix for output files.
+        ext.when   = When to run the module.
+----------------------------------------------------------------------------------------
+*/
+
+
+process {
+    if (params.tools && params.tools.split(',').contains('indexcov')) {
+        withName: 'GOLEFT_INDEXCOV' {
+            ext.when         =  { !params.wes }
+            ext.args         = {""}
+            ext.prefix       = { "indexcov" }
+            publishDir       = [
+                mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/indexcov" },
+                pattern: "*.png"
+            ]
+        }
+    }
+}

conf/modules/indexcov_reindex.config

@@ -0,0 +1,24 @@
+/*
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Config file for defining DSL2 per module options and publishing paths
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Available keys to override module options:
+        ext.args   = Additional arguments appended to command in module.
+        ext.prefix = File name prefix for output files.
+        ext.when   = When to run the module.
+----------------------------------------------------------------------------------------
+*/
+
+
+process {
+    if (params.tools && params.tools.split(',').contains('indexcov')) {
+        withName: 'SAMTOOLS_INDEX_FOR_INDEXCOV' {
+            ext.when         =  { !params.wes }
+            ext.args         = {"-F 3844 -q 30"}
+            ext.prefix       = { "${meta.id}" }
+            publishDir = [
+		enabled : false
+		]
+        }
+    }
+}

conf/test_full_germline.config

@@ -18,6 +18,6 @@ params {
     input = 'https://raw.githubusercontent.com/nf-core/test-datasets/sarek/testdata/csv/NA12878_WGS_30x_full_test.csv'
 
     // Other params
-    tools = 'strelka,freebayes,haplotypecaller,deepvariant,manta,tiddit,cnvkit,vep'
-    split_fastq = 50000000
+    tools = 'strelka,freebayes,haplotypecaller,deepvariant,manta,tiddit,cnvkit,vep,indexcov'
+    split_fastq               = 50000000
 }

modules/local/goleft/indexcov/main.nf

@@ -0,0 +1,51 @@
+process GOLEFT_INDEXCOV {
+label 'process_single'
+
+container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/goleft:0.2.4--h9ee0642_1':
+        'biocontainers/goleft:0.2.4--h9ee0642_1' }"
+
+input:
+        val(meta)
+        path(bams)
+        path(fasta)
+        path(fai)
+output:
+	path("${prefix}/*"),emit:output
+	path("versions.yml"),emit:versions
+script:
+    def args = task.ext.args ?: ''
+    prefix = task.ext.prefix ?: "${meta.id}"
+    def input_files = bams.findAll{it.name.endsWith(".bam")} + bams.findAll{it.name.endsWith(".crai")}
+    def extranormalize = input_files.any{it.name.endsWith(".crai")} ? " --extranormalize " : ""
+"""
+    goleft indexcov \\
+        --fai "${fai}"  \\
+        --directory "${prefix}" \\
+        ${extranormalize} \\
+        $args \\
+        ${input_files.join(" ")}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        goleft: \$(goleft --version 2>&1 | head -n 1 | sed 's/^.*goleft Version: //')
+    END_VERSIONS
+"""
+
+    stub:
+    def args = task.ext.args ?: ''
+    prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    mkdir "${prefix}"
+    touch "${prefix}/${prefix}-indexcov.bed.gz"
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        goleft: \$(goleft --version 2>&1 | head -n 1 | sed 's/^.*goleft Version: //')
+    END_VERSIONS
+    """
+}
+
+
+
+

modules/local/samtools_index_for_indexcov/main.nf

@@ -0,0 +1,57 @@
+process SAMTOOLS_INDEX_FOR_INDEXCOV {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda "bioconda::samtools=1.17"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/samtools:1.17--h00cdaf9_0' :
+        'biocontainers/samtools:1.17--h00cdaf9_0' }"
+
+    input:
+    tuple val(meta), path(input), path(bai)
+    path(fasta)
+    path(fai)
+
+    output:
+    tuple path("${meta.id}.indexcov.bam"),path("${meta.id}.indexcov.bam.bai"), emit: output
+    path  "versions.yml"                 , emit: versions
+
+    script:
+    def args = task.ext.args ?: ''
+
+    """
+    # write BAM header only
+    samtools view --header-only -O BAM  \
+        --threads ${task.cpus} \
+        -o "${meta.id}.indexcov.bam" \
+        --reference "${fasta}" \
+		"${input}"
+
+    # create index without writing BAM (redirecting to /dev/null)
+    samtools view ${args} --uncompressed \
+		--threads ${task.cpus} \
+		-o "/dev/null##idx##${meta.id}.indexcov.bam.bai" \
+		--write-index \
+		-O BAM  \
+	        --reference "${fasta}" \
+		"${input}"
+
+
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        samtools: \$(echo \$(samtools --version 2>&1) | sed 's/^.*samtools //; s/Using.*\$//')
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    touch "${meta.id}.indexcov.bam"
+    touch "${meta.id}.indexcov.bam.bai"
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        samtools: \$(echo \$(samtools --version 2>&1) | sed 's/^.*samtools //; s/Using.*\$//')
+    END_VERSIONS
+    """
+}

nextflow.config

@@ -381,6 +381,8 @@ includeConfig 'conf/modules/cnvkit.config'
 includeConfig 'conf/modules/controlfreec.config'
 includeConfig 'conf/modules/deepvariant.config'
 includeConfig 'conf/modules/freebayes.config'
+includeConfig 'conf/modules/indexcov_reindex.config'
+includeConfig 'conf/modules/goleft.indexcov.config'
 includeConfig 'conf/modules/haplotypecaller.config'
 includeConfig 'conf/modules/joint_germline.config'
 includeConfig 'conf/modules/manta.config'

nextflow_schema.json

@@ -100,7 +100,7 @@
                     "fa_icon": "fas fa-toolbox",
                     "description": "Tools to use for duplicate marking, variant calling and/or for annotation.",
                     "help_text": "Multiple tools separated with commas.\n\n**Variant Calling:**\n\nGermline variant calling can currently be performed with the following variant callers:\n- SNPs/Indels: DeepVariant, FreeBayes, GATK HaplotypeCaller, mpileup, Sentieon Haplotyper, Strelka\n- Structural Variants: Manta, TIDDIT\n- Copy-number: CNVKit\n\nTumor-only somatic variant calling can currently be performed with the following variant callers:\n- SNPs/Indels: FreeBayes, mpileup, Mutect2, Strelka\n- Structural Variants: Manta, TIDDIT\n- Copy-number: CNVKit, ControlFREEC\n\nSomatic variant calling can currently only be performed with the following variant callers:\n- SNPs/Indels: FreeBayes, Mutect2, Strelka2\n- Structural variants: Manta, TIDDIT\n- Copy-Number: ASCAT, CNVKit, Control-FREEC\n- Microsatellite Instability: MSIsensorpro\n\n> **NB** Mutect2 for somatic variant calling cannot be combined with `--no_intervals`\n\n**Annotation:**\n \n- snpEff, VEP, merge (both consecutively).\n\n> **NB** As Sarek will use bgzip and tabix to compress and index VCF files annotated, it expects VCF files to be sorted when starting from `--step annotate`.",
-                    "pattern": "^((ascat|bcfann|cnvkit|controlfreec|deepvariant|freebayes|haplotypecaller|sentieon_dnascope|sentieon_haplotyper|manta|merge|mpileup|msisensorpro|mutect2|sentieon_dedup|snpeff|strelka|tiddit|vep)?,?)*(?<!,)$"
+                    "pattern": "^((ascat|bcfann|cnvkit|controlfreec|deepvariant|freebayes|haplotypecaller|sentieon_dnascope|sentieon_haplotyper|manta|merge|mpileup|msisensorpro|mutect2|sentieon_dedup|snpeff|strelka|tiddit|vep|indexcov)?,?)*(?<!,)$"
                 },
                 "skip_tools": {
                     "type": "string",

subworkflows/local/bam_goleft_indexcov/main.nf

@@ -0,0 +1,29 @@
+//
+// INDEXCOV : call large CNVs using goleft indexcov
+//
+// For all modules here:
+// A when clause condition is defined in the conf/modules.config to determine if the module should be run
+
+include { SAMTOOLS_INDEX_FOR_INDEXCOV  } from '../../../modules/local/samtools_index_for_indexcov/main'
+include { GOLEFT_INDEXCOV              } from '../../../modules/local/goleft/indexcov/main'
+
+
+workflow BAM_GOLEFT_INDEXCOV {
+    take:
+    bam      // channel: [mandatory] [ meta, bam, bai ]
+    fasta     // channel: [mandatory] [ fasta ]
+    fai     // channel: [mandatory] [ fai ]
+
+    main:
+    versions = Channel.empty()
+
+    samtools_index = SAMTOOLS_INDEX_FOR_INDEXCOV(bam, fasta, fai)
+    versions = versions.mix(samtools_index.versions.first())
+
+    indexcov_ch = GOLEFT_INDEXCOV([id:"PREFIX"], samtools_index.output.flatten().collect(), fasta, fai)
+    versions = versions.mix(indexcov_ch.versions)
+
+    emit:
+    versions
+}
+

subworkflows/local/bam_variant_calling_germline_all/main.nf

@@ -16,6 +16,7 @@ include { BAM_VARIANT_CALLING_SINGLE_STRELKA      } from '../bam_variant_calling
 include { BAM_VARIANT_CALLING_SINGLE_TIDDIT       } from '../bam_variant_calling_single_tiddit/main'
 include { SENTIEON_DNAMODELAPPLY                  } from '../../../modules/nf-core/sentieon/dnamodelapply/main'
 include { VCF_VARIANT_FILTERING_GATK              } from '../vcf_variant_filtering_gatk/main'
+include { BAM_GOLEFT_INDEXCOV                     } from '../bam_goleft_indexcov/main'
 
 
 
@@ -190,6 +191,17 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
         versions = versions.mix(BAM_VARIANT_CALLING_GERMLINE_MANTA.out.versions)
     }
 
+    // INDEXCOV
+    if (tools.split(',').contains('indexcov')) {
+        BAM_GOLEFT_INDEXCOV (
+            cram,
+            fasta,
+            fasta_fai
+        )
+     	versions = versions.mix(BAM_GOLEFT_INDEXCOV.out.versions)
+    }
+
+
     // SENTIEON DNASCOPE
     if (tools.split(',').contains('sentieon_dnascope')) {
         BAM_VARIANT_CALLING_SENTIEON_DNASCOPE(