Merge pull request #393 from genomic-medicine-sweden/sortvcf

sort ranked SVs before indexing
nf-core · Aug 8, 2023 · b0e4832 · b0e4832
2 parents ce03dfa + b946556
commit b0e4832
Show file tree

Hide file tree

Showing 11 changed files with 277 additions and 11 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -17,6 +17,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Invalid GATK4 container which caused incorrect singularity downloads with nf-core download [nf-core/modules #3668](https://github.com/nf-core/modules/issues/3668)
 - Make the default cram prefix same as markduplicates prefix [#392](https://github.com/nf-core/raredisease/pull/392)
+- Sort ranked SV vcf before indexing with tabix [#393](https://github.com/nf-core/raredisease/pull/393)
 - Make target bed file optional for WGS mode (Issue [#375](https://github.com/nf-core/raredisease/issues/375)) [#395](https://github.com/nf-core/raredisease/pull/395)
 - Added constraints to block the pipeline from running CollectWgsMetrics on WES samples [#396](https://github.com/nf-core/raredisease/pull/396)
 

diff --git a/conf/modules/rank_variants.config b/conf/modules/rank_variants.config
@@ -37,14 +37,27 @@ process {
         ext.prefix = { "${meta.id}_ranked_sv" }
     }
 
-    withName: '.*RANK_VARIANTS_SV:TABIX_BGZIPTABIX' {
+    withName: '.*RANK_VARIANTS_SV:BCFTOOLS_SORT' {
+        ext.args = "--output-type z"
         ext.prefix = { "${meta.id}_ranked_sv" }
         publishDir = [
             path: { "${params.outdir}/rank_and_filter" },
             mode: params.publish_dir_mode,
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
         ]
     }
+
+    withName: '.*RANK_VARIANTS_SV:TABIX_BGZIP' {
+        ext.when = false
+    }
+
+    withName: '.*RANK_VARIANTS_SV:TABIX_TABIX' {
+        publishDir = [
+            path: { "${params.outdir}/rank_and_filter" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
 }
 
 //
@@ -77,12 +90,24 @@ process {
         ext.prefix = { "${meta.id}_ranked_snv" }
     }
 
-    withName: '.*RANK_VARIANTS_SNV:TABIX_BGZIPTABIX' {
+    withName: '.*RANK_VARIANTS_SNV:BCFTOOLS_SORT' {
+        ext.when = false
+    }
+
+    withName: '.*RANK_VARIANTS_SNV:TABIX_BGZIP' {
         ext.prefix = { "${meta.id}_ranked_snv" }
         publishDir = [
             path: { "${params.outdir}/rank_and_filter" },
             mode: params.publish_dir_mode,
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
         ]
     }
+
+    withName: '.*RANK_VARIANTS_SNV:TABIX_TABIX' {
+        publishDir = [
+            path: { "${params.outdir}/rank_and_filter" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
 }
diff --git a/conf/modules/raredisease.config b/conf/modules/raredisease.config
@@ -105,6 +105,10 @@ process {
 process {
     withName: '.*FILTER_VEP_SNV' {
         ext.prefix = { "${meta.id}_clinical_snv" }
+    }
+
+    withName: '.*BGZIPTABIX_SNV' {
+        ext.prefix = { "${meta.id}_clinical_snv" }
         publishDir = [
             path: { "${params.outdir}/rank_and_filter" },
             mode: params.publish_dir_mode,
@@ -116,6 +120,10 @@ process {
 process {
     withName: '.*FILTER_VEP_SV' {
         ext.prefix = { "${meta.id}_clinical_sv" }
+    }
+
+    withName: '.*BGZIPTABIX_SV' {
+        ext.prefix = { "${meta.id}_clinical_sv" }
         publishDir = [
             path: { "${params.outdir}/rank_and_filter" },
             mode: params.publish_dir_mode,

diff --git a/modules.json b/modules.json
@@ -40,6 +40,11 @@
                         "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
                         "installed_by": ["modules"]
                     },
+                    "bcftools/sort": {
+                        "branch": "master",
+                        "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
+                        "installed_by": ["modules"]
+                    },
                     "bcftools/view": {
                         "branch": "master",
                         "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
@@ -360,6 +365,11 @@
                         "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
                         "installed_by": ["modules"]
                     },
+                    "tabix/bgzip": {
+                        "branch": "master",
+                        "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
+                        "installed_by": ["modules"]
+                    },
                     "tabix/bgziptabix": {
                         "branch": "master",
                         "git_sha": "591b71642820933dcb3c954c934b397bd00d8e5e",

diff --git a/modules/local/filter_vep.nf b/modules/local/filter_vep.nf
@@ -13,8 +13,8 @@ process FILTER_VEP {
     path (select_feature_file)
 
     output:
-    tuple val(meta), path("*.ann_filter.vcf.gz"), emit: vcf
-    path "versions.yml"                         , emit: versions
+    tuple val(meta), path("*.ann_filter.vcf"), emit: vcf
+    path "versions.yml"                      , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
@@ -26,7 +26,7 @@ process FILTER_VEP {
     filter_vep \\
         --format vcf \\
         --input_file $vcf \\
-        --output_file ${prefix}.ann_filter.vcf.gz \\
+        --output_file ${prefix}.ann_filter.vcf \\
         --only_matched \\
         --filter \"HGNC_ID in ${select_feature_file}\"
 
@@ -39,7 +39,7 @@ process FILTER_VEP {
     stub:
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
-    touch ${prefix}.ann_filter.vcf.gz
+    touch ${prefix}.ann_filter.vcf
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

diff --git a/modules/nf-core/bcftools/sort/main.nf b/modules/nf-core/bcftools/sort/main.nf
diff --git a/modules/nf-core/bcftools/sort/meta.yml b/modules/nf-core/bcftools/sort/meta.yml
diff --git a/modules/nf-core/tabix/bgzip/main.nf b/modules/nf-core/tabix/bgzip/main.nf
diff --git a/modules/nf-core/tabix/bgzip/meta.yml b/modules/nf-core/tabix/bgzip/meta.yml
diff --git a/subworkflows/local/rank_variants.nf b/subworkflows/local/rank_variants.nf
@@ -6,7 +6,9 @@ include { GENMOD_ANNOTATE  } from '../../modules/nf-core/genmod/annotate/main'
 include { GENMOD_MODELS    } from '../../modules/nf-core/genmod/models/main'
 include { GENMOD_SCORE     } from '../../modules/nf-core/genmod/score/main'
 include { GENMOD_COMPOUND  } from '../../modules/nf-core/genmod/compound/main'
-include { TABIX_BGZIPTABIX } from '../../modules/nf-core/tabix/bgziptabix/main'
+include { BCFTOOLS_SORT    } from '../../modules/nf-core/bcftools/sort/main'
+include { TABIX_BGZIP      } from '../../modules/nf-core/tabix/bgzip/main'
+include { TABIX_TABIX      } from '../../modules/nf-core/tabix/tabix/main'
 
 workflow RANK_VARIANTS {
 
@@ -27,15 +29,23 @@ workflow RANK_VARIANTS {
 
         GENMOD_COMPOUND(GENMOD_SCORE.out.vcf)
 
-        TABIX_BGZIPTABIX (GENMOD_COMPOUND.out.vcf)
+        BCFTOOLS_SORT(GENMOD_COMPOUND.out.vcf) // SV file needs to be sorted before indexing
+
+        TABIX_BGZIP(GENMOD_COMPOUND.out.vcf) //run only for SNVs
+
+        ch_vcf = TABIX_BGZIP.out.output.mix(BCFTOOLS_SORT.out.vcf)
+
+        TABIX_TABIX (ch_vcf)
 
         ch_versions = ch_versions.mix(GENMOD_ANNOTATE.out.versions)
         ch_versions = ch_versions.mix(GENMOD_MODELS.out.versions)
         ch_versions = ch_versions.mix(GENMOD_SCORE.out.versions)
         ch_versions = ch_versions.mix(GENMOD_COMPOUND.out.versions)
-        ch_versions = ch_versions.mix(TABIX_BGZIPTABIX.out.versions)
+        ch_versions = ch_versions.mix(BCFTOOLS_SORT.out.versions)
+        ch_versions = ch_versions.mix(TABIX_BGZIP.out.versions)
+        ch_versions = ch_versions.mix(TABIX_TABIX.out.versions)
 
     emit:
-        vcf      = TABIX_BGZIPTABIX.out.gz_tbi.map { meta, vcf, tbi -> return [ meta, vcf ] }.collect() // channel: [ val(meta), path(vcf) ]
-        versions = ch_versions                                                                          // channel: [ path(versions.yml) ]
+        vcf      = ch_vcf       // channel: [ val(meta), path(vcf) ]
+        versions = ch_versions  // channel: [ path(versions.yml) ]
 }