Add filtering with bcftools

CHANGELOG.md

@@ -9,6 +9,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- [2044](https://github.com/nf-core/sarek/pull/2044) - Add filtering with `bcftools view -f PASS` following variantcalling step
+
 ### Changed
 
 ### Fixed
@@ -64,8 +66,10 @@ This patch release includes a bump to Nextflow 25.04.8.
 
 ### Parameters
 
-| Params | status |
-| ------ | ------ |
+| Params                       | status |
+| ---------------------------- | ------ |
+| `--filter_vcfs`              | New    |
+| `--bcftools_filter_criteria` | New    |
 
 ### Developer section
 

README.md

@@ -67,7 +67,7 @@ Depending on the options and samples provided, the pipeline can currently perfor
   - `Strelka`
   - `TIDDIT`
 - Post-variant calling options, one of:
-  - `BCFtools concat` for germline vcfs and/or `BCFtools norm` for all vcfs (_experimental Feature_)
+  - `BCFtools concat` for germline vcfs and/or `BCFtools view` (default: filter by `PASS`), `BCFtools norm` for all vcfs
   - `Varlociraptor` for all vcfs
 - Variant filtering and annotation (`SnpEff`, `Ensembl VEP`, `BCFtools annotate`)
 - Summarise and represent QC (`MultiQC`)

conf/modules/freebayes.config

@@ -62,6 +62,14 @@ process {
         ]
     }
 
+    withName: 'TABIX_VC_FREEBAYES_FILT' {
+        publishDir = [
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/variant_calling/freebayes/${meta.id}/" },
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+
     // PAIR_VARIANT_CALLING
     if (params.tools && params.tools.split(',').contains('freebayes')) {
         withName: '.*:BAM_VARIANT_CALLING_SOMATIC_ALL:BAM_VARIANT_CALLING_FREEBAYES:FREEBAYES' {

conf/modules/post_variant_calling.config

@@ -16,34 +16,29 @@
 
 process {
 
-    withName: 'GERMLINE_VCFS_CONCAT' {
-        ext.args   = { "-a" }
-        publishDir = [ enabled: false ]
+    withName: 'FILTER_VCFS' {
+        ext.args = { [ params.bcftools_filter_criteria,
+                        "--output-type z --write-index=tbi"
+                    ].join(" ").trim() }
+        ext.prefix = { vcf.baseName - '.vcf' + '.bcftools_filtered' }
+        publishDir = [
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/variant_calling/filtered/${meta.id}/" },
+            pattern: "*{.tbi,.vcf.gz}"
+        ]
     }
 
     withName: 'ADD_INFO_TO_VCF' {
         publishDir = [ enabled: false ]
     }
 
-    withName: 'GERMLINE_VCFS_CONCAT_SORT' {
-        ext.prefix = { "${meta.id}.germline" }
-        publishDir = [
-            mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" },
-                pattern: "*vcf.gz"
-        ]
-    }
-
-    withName: 'VCFS_NORM_SORT' {
-        ext.prefix = { "${meta.id}.${meta.variantcaller}.norm" }
-        publishDir = [
-            mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" },
-                pattern: "*vcf.gz"
-        ]
+    withName: 'TABIX_EXT_VCF' {
+        ext.prefix = { "${input.baseName}" }
+        publishDir = [ enabled: false ]
     }
 
     withName: 'VCFS_NORM' {
+        ext.prefix = { vcf.baseName - '.added_info.vcf' + '.norm' }
         ext.args = { [
             '--multiallelics -both', // split multiallelic sites into biallelic records and both SNPs and indels should be merged separately into two records
             '--rm-dup all'           // output only the first instance of a record which is present multiple times
@@ -53,27 +48,29 @@ process {
         ]
     }
 
-    withName: 'TABIX_EXT_VCF' {
-        ext.prefix = { "${input.baseName}" }
-        publishDir = [ enabled: false ]
-    }
-
-    withName: 'TABIX_GERMLINE_VCFS_CONCAT_SORT' {
-        ext.prefix = { "${meta.id}.germline" }
+    withName: 'VCFS_NORM_SORT' {
+        ext.prefix = { vcf.baseName - '.vcf' + '.sorted' }
+        ext.args   = "--output-type z --write-index=tbi"
+        ext.when   = { params.normalize_vcfs }
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" },
-                pattern: "*.tbi"
+                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" },
+                pattern: "*{vcf.gz,tbi}"
         ]
     }
 
-    withName: 'TABIX_VCFS_NORM_SORT' {
-        ext.prefix = { "${meta.id}.${meta.variantcaller}.norm" }
+    withName: 'GERMLINE_VCFS_CONCAT' {
+        ext.args   = { "-a --output-type z" }
+        publishDir = [ enabled: false ]
+    }
+
+    withName: 'GERMLINE_VCFS_CONCAT_SORT' {
+        ext.prefix = { "${meta.id}.germline" }
+        ext.args   = "--output-type z --write-index=tbi"
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" },
-                pattern: "*.tbi"
+                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" },
+                pattern: "*{vcf.gz,tbi}"
         ]
     }
-
 }

conf/test_full.config

@@ -22,6 +22,7 @@ params {
     split_fastq = 20000000
     intervals   = 's3://ngi-igenomes/test-data/sarek/S07604624_Padded_Agilent_SureSelectXT_allexons_V6_UTR.bed'
     wes         = true
+    filter_vcfs = true
 }
 
 process {

conf/test_full_germline.config

@@ -20,4 +20,5 @@ params {
     // Other params
     tools       = 'cnvkit,deepvariant,freebayes,haplotypecaller,indexcov,manta,strelka,tiddit,snpeff,vep'
     split_fastq = 50000000
+    filter_vcfs = true
 }