Sarek bcftools normalization

CHANGELOG.md

@@ -9,7 +9,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
-- [1759](https://github.com/nf-core/sarek/pull/1759) - Back to dev
+- [1682](https://github.com/nf-core/sarek/pull/1682) - Add `bcftools_norm` in `POST_VARIANTCALLING` for normalization of all vcf files; edit vcf_concatenate_germline subworkflow
+- [1760](https://github.com/nf-core/sarek/pull/1760) - Back to dev
 
 ### Changed
 

README.md

@@ -62,6 +62,7 @@ Depending on the options and samples provided, the pipeline can currently perfor
   - `Strelka2`
   - `TIDDIT`
   - `Lofreq`
+- Post-variant calling options (`BCFtools concat` for germline vcfs, `BCFtools norm` for all vcfs)
 - Variant filtering and annotation (`SnpEff`, `Ensembl VEP`, `BCFtools annotate`)
 - Summarise and represent QC (`MultiQC`)
 
@@ -183,6 +184,7 @@ We thank the following people for their extensive assistance in the development
 - [Szilveszter Juhos](https://github.com/szilvajuhos)
 - [Tobias Koch](https://github.com/KochTobi)
 - [Winni Kretzschmar](https://github.com/winni2k)
+- [Patricie Skaláková](https://github.com/Patricie34)
 
 ## Acknowledgements
 

conf/modules/post_variant_calling.config

@@ -16,7 +16,7 @@
 
 process {
 
-    withName: 'GERMLINE_VCFS_CONCAT'{
+    withName: 'GERMLINE_VCFS_CONCAT' {
         ext.args   = { "-a" }
         ext.when   = { params.concatenate_vcfs }
         publishDir = [
@@ -25,26 +25,61 @@ process {
         ]
     }
 
-    withName: 'GERMLINE_VCFS_CONCAT_SORT'{
+    withName: 'GERMLINE_VCFS_CONCAT_SORT' {
         ext.prefix = { "${meta.id}.germline" }
         ext.when   = { params.concatenate_vcfs }
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" },
+                pattern: "*vcf.gz"
+        ]
+    }
+
+    withName: 'VCFS_NORM_SORT' {
+        ext.prefix = { "${meta.id}.${meta.variantcaller}.norm" }
+        ext.when   = { params.normalize_vcfs }
+        publishDir = [
+            mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" },
+                pattern: "*vcf.gz"
+        ]
+    }
+
+    withName: 'VCFS_NORM' {
+        ext.args = { [
+            '--multiallelics -both', //split multiallelic sites into biallelic records and both SNPs and indels should be merged separately into two records
+            '--rm-dup all'            //output only the first instance of a record which is present multiple times
+        ].join(' ') }
+        ext.when   = { params.normalize_vcfs }
+        publishDir = [
+            mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" }
         ]
     }
 
     withName: 'TABIX_EXT_VCF' {
-        ext.prefix = { "${input.baseName}" }
-        ext.when   = { params.concatenate_vcfs }
+    ext.prefix = { "${input.baseName}" }
+    ext.when = { params.concatenate_vcfs || params.normalize_vcfs }
     }
 
     withName: 'TABIX_GERMLINE_VCFS_CONCAT_SORT'{
         ext.prefix = { "${meta.id}.germline" }
         ext.when   = { params.concatenate_vcfs }
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" },
+                pattern: "*.tbi"
+        ]
+    }
+
+    withName: 'TABIX_VCFS_NORM_SORT'{
+        ext.prefix = { "${meta.id}.${meta.variantcaller}.norm" }
+        ext.when   = { params.normalize_vcfs }
+        publishDir = [
+            mode: params.publish_dir_mode,
+                path: { "${params.outdir}/variant_calling/normalized/${meta.id}/" },
+                pattern: "*.tbi"
         ]
     }
 }
+