Sarek bcftools normalization

CHANGELOG.md

@@ -13,8 +13,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [1642](https://github.com/nf-core/sarek/pull/1642) - Back to dev
 - [1653](https://github.com/nf-core/sarek/pull/1653) - Updates `sarek_subway` files with `lofreq`
 - [1660](https://github.com/nf-core/sarek/pull/1642) - Add `--length_required` for minimal reads length with `FASTP`
-- [1663](https://github.com/nf-core/sarek/pull/1663) - Massive conda modules update
 - [1664](https://github.com/nf-core/sarek/pull/1664) - Check if flowcell ID matches for read pair
+- [1663](https://github.com/nf-core/sarek/pull/1663) - Massive conda modules update
+- [1680](https://github.com/nf-core/sarek/pull/1682) - Add `bcftools_norm` in `POST_VARIANTCALLING` for normalization of all vcf files or for concatenated germline vcfs
 
 ### Changed
 

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 +25,39 @@ process {
         ]
     }
 
-    withName: 'GERMLINE_VCFS_CONCAT_SORT'{
-        ext.prefix = { "${meta.id}.germline" }
-        ext.when   = { params.concatenate_vcfs }
+    withName: 'GERMLINE_VCFS_CONCAT_SORT|VCFS_NORM_SORT' {
+        ext.prefix = { "${meta.id}.${processName == 'GERMLINE_VCFS_CONCAT_SORT' ? 'germline' : 'norm'}" }
+        ext.when   = { params.concatenate_vcfs || params.normalize_vcfs }
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+                path: { "${params.outdir}/variant_calling/${processName == 'GERMLINE_VCFS_CONCAT_SORT' ? 'concat' : 'normalized'}/${meta.id}/" }
+        ]
+    }
+
+    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 = { "${meta.id}.germline" }
-        ext.when   = { params.concatenate_vcfs }
+    withName: 'TABIX_GERMLINE_VCFS_CONCAT_SORT|TABIX_VCFS_NORM_SORT' {
+        ext.prefix = { "${meta.id}.${processName == 'TABIX_GERMLINE_VCFS_CONCAT_SORT' ? 'germline' : 'norm'}" }
+        ext.when   = { params.concatenate_vcfs || params.normalize_vcfs }
         publishDir = [
             mode: params.publish_dir_mode,
-                path: { "${params.outdir}/variant_calling/concat/${meta.id}/" }
+                path: { "${params.outdir}/variant_calling/${processName == 'TABIX_GERMLINE_VCFS_CONCAT_SORT' ? 'concat' : 'normalized'}/${meta.id}/" }
         ]
     }
 }
+

modules.json

@@ -26,6 +26,11 @@
                         "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",
                         "installed_by": ["bam_ngscheckmate"]
                     },
+                    "bcftools/norm": {
+                        "branch": "master",
+                        "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",
+                        "installed_by": ["modules"]
+                    },
                     "bcftools/sort": {
                         "branch": "master",
                         "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",