Concatenating germline vcfs

conf/modules/modules.config

@@ -87,6 +87,24 @@ process {
         }
     }
 
+    // CONCATENATED, SORT, UNANNOTATED VCFS
+    withName: 'GERMLINE_VCFS_CONCAT_SORT'{
+        ext.prefix       = { "${meta.id}.germline" }
+        publishDir       = [
+            mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+        ]
+    }
+
+    withName: 'TABIX_GERMLINE_VCFS_CONCAT_SORT'{
+        ext.prefix       = { "${meta.id}.germline" }
+        publishDir       = [
+            mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+        ]
+    }
+
+
     // VCF
     withName: 'BCFTOOLS_STATS' {
         ext.when         = { !(params.skip_tools && params.skip_tools.split(',').contains('bcftools')) }

modules.json

@@ -15,6 +15,10 @@
                         "git_sha": "6301e29d77e7ec7ce98b55b8a361b316a9a91bfe",
                         "installed_by": ["modules"]
                     },
+                    "bcftools/concat": {
+                        "branch": "master",
+                        "git_sha": "5e34754d42cd2d5d248ca8673c0a53cdf5624905"
+                    },
                     "bcftools/sort": {
                         "branch": "master",
                         "git_sha": "78cf39939fbe160a1410c44a6c5946f9a4c56e7e",

nextflow.config

@@ -63,6 +63,7 @@ params {
     cf_minqual                  = 0     // ControlFreec default values
     cf_window                   = null  // by default we are not using this in Control-FREEC
     cnvkit_reference            = null  // by default the reference is build from the fasta file
+    concatenate_vcfs            = false // by default we don't concatenate the germline-vcf-files
     ignore_soft_clipped_bases   = false // no --dont-use-soft-clipped-bases for GATK Mutect2
     wes                         = false // Set to true, if data is exome/targeted sequencing data. Used to use correct models in various variant callers
     joint_germline              = false // g.vcf & joint germline calling are not run by default if HaplotypeCaller is selected

nextflow_schema.json

@@ -226,6 +226,12 @@
             "default": "",
             "fa_icon": "fas fa-toolbox",
             "properties": {
+                "concatenate_vcfs": {
+                    "type": "boolean",
+                    "fa_icon": "fas fa-merge",
+                    "description": "Option for concatenating germline vcf-files.",
+                    "help_text": "Concatenating the germline vcf-files from each applied variant-caller into one vcf-file using bfctools concat."
+                },
                 "only_paired_variant_calling": {
                     "type": "boolean",
                     "fa_icon": "fas fa-forward",

subworkflows/local/bam_variant_calling_deepvariant/main.nf

@@ -23,6 +23,11 @@ workflow BAM_VARIANT_CALLING_DEEPVARIANT {
         no_intervals: it[0].num_intervals <= 1
     }.set{deepvariant_vcf_out}
 
+    DEEPVARIANT.out.vcf_tbi.branch{
+        intervals:    it[0].num_intervals > 1
+        no_intervals: it[0].num_intervals <= 1
+    }.set{deepvariant_tbi_out}
+
     DEEPVARIANT.out.gvcf.branch{
         intervals:    it[0].num_intervals > 1
         no_intervals: it[0].num_intervals <= 1
@@ -98,6 +103,22 @@ workflow BAM_VARIANT_CALLING_DEEPVARIANT {
                         ], vcf]
                     }
 
+    deepvariant_vcf_tbi = Channel.empty().mix(
+                        MERGE_DEEPVARIANT_VCF.out.tbi,
+                        deepvariant_tbi_out.no_intervals)
+                    .map{ meta, tbi ->
+                        [[
+                            id:             meta.sample,
+                            num_intervals:  meta.num_intervals,
+                            patient:        meta.patient,
+                            sample:         meta.sample,
+                            sex:            meta.sex,
+                            status:         meta.status,
+                            variantcaller:  "deepvariant"
+                        ], tbi]
+                    }
+
+
     ch_versions = ch_versions.mix(MERGE_DEEPVARIANT_GVCF.out.versions)
     ch_versions = ch_versions.mix(MERGE_DEEPVARIANT_VCF.out.versions)
     ch_versions = ch_versions.mix(DEEPVARIANT.out.versions)
@@ -106,6 +127,7 @@ workflow BAM_VARIANT_CALLING_DEEPVARIANT {
 
     emit:
     deepvariant_vcf
+    deepvariant_vcf_tbi
     deepvariant_gvcf
     versions = ch_versions
 }

subworkflows/local/bam_variant_calling_freebayes/main.nf

@@ -72,12 +72,29 @@ workflow BAM_VARIANT_CALLING_FREEBAYES {
                             vcf]
                     }
 
+    freebayes_vcf_tbi  = Channel.empty().mix(
+                            MERGE_FREEBAYES.out.tbi,
+                            TABIX_VC_FREEBAYES.out.tbi)
+                        .map{ meta, tbi ->
+                            [ [
+                                id:             meta.id,
+                                normal_id:      meta.normal_id,
+                                num_intervals:  meta.num_intervals,
+                                patient:        meta.patient,
+                                sex:            meta.sex,
+                                tumor_id:       meta.tumor_id,
+                                variantcaller:  "freebayes"
+                            ],
+                                tbi]
+                        }
+
     ch_versions = ch_versions.mix(BCFTOOLS_SORT.out.versions)
     ch_versions = ch_versions.mix(MERGE_FREEBAYES.out.versions)
     ch_versions = ch_versions.mix(FREEBAYES.out.versions)
     ch_versions = ch_versions.mix(TABIX_VC_FREEBAYES.out.versions)
 
     emit:
     freebayes_vcf
+    freebayes_vcf_tbi
     versions = ch_versions
 }

subworkflows/local/bam_variant_calling_germline_all/main.nf

@@ -37,13 +37,20 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
     //TODO: Temporary until the if's can be removed and printing to terminal is prevented with "when" in the modules.config
     deepvariant_vcf     = Channel.empty()
     freebayes_vcf       = Channel.empty()
-    genotype_gvcf       = Channel.empty()
     haplotypecaller_vcf = Channel.empty()
     manta_vcf           = Channel.empty()
     mpileup_vcf         = Channel.empty()
     strelka_vcf         = Channel.empty()
     tiddit_vcf          = Channel.empty()
 
+    deepvariant_vcf_tbi     = Channel.empty()
+    freebayes_vcf_tbi       = Channel.empty()
+    haplotypecaller_vcf_tbi = Channel.empty()
+    manta_vcf_tbi           = Channel.empty()
+    mpileup_vcf_tbi         = Channel.empty()
+    strelka_vcf_tbi         = Channel.empty()
+    tiddit_vcf_tbi          = Channel.empty()
+
     // Remap channel with intervals
     cram_recalibrated_intervals = cram_recalibrated.combine(intervals)
         .map{ meta, cram, crai, intervals, num_intervals ->
@@ -95,8 +102,8 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             dict
         )
 
-        mpileup_germline = BAM_VARIANT_CALLING_MPILEUP.out.mpileup
         mpileup_vcf = BAM_VARIANT_CALLING_MPILEUP.out.vcf
+        mpileup_vcf_tbi = BAM_VARIANT_CALLING_MPILEUP.out.tbi
         ch_versions = ch_versions.mix(BAM_VARIANT_CALLING_MPILEUP.out.versions)
     }
 
@@ -116,7 +123,7 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             []
         )
 
-        ch_versions     = ch_versions.mix(BAM_VARIANT_CALLING_CNVKIT.out.versions)
+        ch_versions = ch_versions.mix(BAM_VARIANT_CALLING_CNVKIT.out.versions)
     }
 
     // DEEPVARIANT
@@ -128,8 +135,9 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             fasta_fai
         )
 
-        deepvariant_vcf = Channel.empty().mix(BAM_VARIANT_CALLING_DEEPVARIANT.out.deepvariant_vcf)
-        ch_versions     = ch_versions.mix(BAM_VARIANT_CALLING_DEEPVARIANT.out.versions)
+        deepvariant_vcf      = Channel.empty().mix(BAM_VARIANT_CALLING_DEEPVARIANT.out.deepvariant_vcf)
+        deepvariant_vcf_tbi  = Channel.empty().mix(BAM_VARIANT_CALLING_DEEPVARIANT.out.deepvariant_vcf_tbi)
+        ch_versions          = ch_versions.mix(BAM_VARIANT_CALLING_DEEPVARIANT.out.versions)
     }
 
     // FREEBAYES
@@ -147,8 +155,9 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             fasta_fai
         )
 
-        freebayes_vcf   = BAM_VARIANT_CALLING_FREEBAYES.out.freebayes_vcf
-        ch_versions     = ch_versions.mix(BAM_VARIANT_CALLING_FREEBAYES.out.versions)
+        freebayes_vcf     = BAM_VARIANT_CALLING_FREEBAYES.out.freebayes_vcf
+        freebayes_vcf_tbi = BAM_VARIANT_CALLING_FREEBAYES.out.freebayes_vcf_tbi
+        ch_versions       = ch_versions.mix(BAM_VARIANT_CALLING_FREEBAYES.out.versions)
     }
 
     // HAPLOTYPECALLER
@@ -184,8 +193,10 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
                         known_sites_snps_tbi,
                         intervals_bed_combined_haplotypec)
 
-        haplotypecaller_vcf  = BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.filtered_vcf
-        ch_versions          = ch_versions.mix(BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.versions)
+        haplotypecaller_vcf      = BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.filtered_vcf
+        haplotypecaller_vcf_tbi  = BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.filtered_vcf_tbi
+        ch_versions              = ch_versions.mix(BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.versions)
+
     }
 
     // MANTA
@@ -197,8 +208,10 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             fasta_fai
         )
 
-        manta_vcf   = BAM_VARIANT_CALLING_GERMLINE_MANTA.out.manta_vcf
-        ch_versions = ch_versions.mix(BAM_VARIANT_CALLING_GERMLINE_MANTA.out.versions)
+
+        manta_vcf     = BAM_VARIANT_CALLING_GERMLINE_MANTA.out.manta_vcf
+        manta_vcf_tbi = BAM_VARIANT_CALLING_GERMLINE_MANTA.out.manta_vcf_tbi
+        ch_versions   = ch_versions.mix(BAM_VARIANT_CALLING_GERMLINE_MANTA.out.versions)
     }
 
     // STRELKA
@@ -210,8 +223,9 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             fasta_fai
         )
 
-        strelka_vcf = BAM_VARIANT_CALLING_SINGLE_STRELKA.out.strelka_vcf
-        ch_versions = ch_versions.mix(BAM_VARIANT_CALLING_SINGLE_STRELKA.out.versions)
+        strelka_vcf     = BAM_VARIANT_CALLING_SINGLE_STRELKA.out.strelka_vcf
+        strelka_vcf_tbi = BAM_VARIANT_CALLING_SINGLE_STRELKA.out.strelka_vcf_tbi
+        ch_versions     = ch_versions.mix(BAM_VARIANT_CALLING_SINGLE_STRELKA.out.versions)
     }
 
     //TIDDIT
@@ -222,19 +236,27 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
             bwa
         )
 
-        tiddit_vcf = BAM_VARIANT_CALLING_SINGLE_TIDDIT.out.tiddit_vcf
-        ch_versions = ch_versions.mix(BAM_VARIANT_CALLING_SINGLE_TIDDIT.out.versions)
+        tiddit_vcf     = BAM_VARIANT_CALLING_SINGLE_TIDDIT.out.tiddit_vcf
+        tiddit_vcf_tbi = BAM_VARIANT_CALLING_SINGLE_TIDDIT.out.tiddit_vcf_tbi
+        ch_versions    = ch_versions.mix(BAM_VARIANT_CALLING_SINGLE_TIDDIT.out.versions)
     }
 
     emit:
     deepvariant_vcf
     freebayes_vcf
-    genotype_gvcf
     haplotypecaller_vcf
     manta_vcf
     mpileup_vcf
     strelka_vcf
     tiddit_vcf
 
+    deepvariant_vcf_tbi
+    freebayes_vcf_tbi
+    haplotypecaller_vcf_tbi
+    manta_vcf_tbi
+    mpileup_vcf_tbi
+    strelka_vcf_tbi
+    tiddit_vcf_tbi
+
     versions = ch_versions
 }