Merge pull request #557 from FriederikeHanssen/single_sample_gatk

Haplotypecaller fix + single sample germline VC
nf-core · Jun 16, 2022 · 6bfbdb1 · 6bfbdb1
2 parents a3d836b + cc5fd50
commit 6bfbdb1
Show file tree

Hide file tree

Showing 16 changed files with 471 additions and 101 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -27,6 +27,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#538](https://github.com/nf-core/sarek/pull/538) - Add param `--seq_platform`, default: `ILLUMINA`
 - [#545](https://github.com/nf-core/sarek/pull/545) - Add modules and subworkflows for `cnvkit` tumor_only mode
 - [#540](https://github.com/nf-core/sarek/pull/540) - Add modules and subworkflows for `cnvkit` somatic mode
+- [#557](https://github.com/nf-core/sarek/pull/557) - Add `Haplotypecaller` single sample mode together with `CNNScoreVariants` and `FilterVariantTranches`
 - [#576](https://github.com/nf-core/sarek/pull/576) - Add modules and subworkflows for `cnvkit` germline mode
 
 ### Changed

diff --git a/conf/modules.config b/conf/modules.config
@@ -111,7 +111,7 @@ process {
     }
 
     withName: 'TABIX_KNOWN_INDELS' {
-        ext.when         = { !params.known_indels_tbi && params.known_indels && (params.step == 'mapping' || params.step == "markduplicates" || params.step == 'prepare_recalibration') }
+        ext.when         = { !params.known_indels_tbi && params.known_indels && (params.step == 'mapping' || params.step == "markduplicates" || params.step == 'prepare_recalibration' || (params.tools && params.tools.contains('haplotypecaller')) ) }
         publishDir       = [
             enabled: params.save_reference,
             mode: params.publish_dir_mode,
@@ -549,18 +549,20 @@ process{
     }
 
     // HAPLOTYPECALLER
-    withName: 'MERGE_HAPLOTYPECALLER' {
-        ext.prefix       = {"${meta.id}.g"}
+    withName: 'MERGE_HAPLOTYPECALLER.*' {
+        ext.prefix       = { params.joint_germline ? "${meta.id}.g" : "${meta.id}" }
         publishDir       = [
-            enabled: !params.no_intervals,
             mode: params.publish_dir_mode,
             path: { "${params.outdir}/variant_calling/${meta.id}/haplotypecaller" },
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
         ]
     }
+    withName: 'MERGE_HAPLOTYPECALLER_FILTERED' {
+        ext.prefix       = { "${meta.id}.filtered" }
+    }
     withName: 'HAPLOTYPECALLER' {
         ext.args         = { params.joint_germline ? "-ERC GVCF" : "" }
-        ext.prefix       = { meta.num_intervals <= 1 ? "${meta.id}.g" : "${meta.id}_${intervals.simpleName}.g" }
+        ext.prefix       = { meta.num_intervals <= 1 ? ( params.joint_germline ? "${meta.id}.g" : "${meta.id}" ) : ( params.joint_germline ? "${meta.id}_${intervals.simpleName}.g" :"${meta.id}_${intervals.simpleName}" ) }
         ext.when         = { params.tools && params.tools.contains('haplotypecaller') }
         publishDir       = [
             mode: params.publish_dir_mode,
@@ -569,10 +571,26 @@ process{
             saveAs: { meta.num_intervals > 1 ? null : "haplotypecaller/${it}" }
         ]
     }
+    withName: 'CNNSCOREVARIANTS' {
+        ext.prefix       = { meta.num_intervals <= 1 ?  "${meta.id}" : "${meta.id}_${intervals.simpleName}" }
+        publishDir       = [
+            enabled: false
+        ]
+    }
+    withName: 'FILTERVARIANTTRANCHES' {
+        ext.args         = { "--info-key CNN_1D" }
+        ext.prefix       = { meta.num_intervals <= 1 ? "${meta.id}" : "${meta.id}_${intervals.simpleName}" }
+        publishDir       = [
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/variant_calling/${meta.id}/"},
+            pattern: "*{vcf.gz,vcf.gz.tbi}",
+            saveAs: { meta.num_intervals > 1 ? null : "haplotypecaller/${it}" }
+        ]
+    }
+
     withName: 'GENOTYPEGVCFS' {
         ext.when         = { params.tools && params.tools.contains('haplotypecaller') && params.joint_germline}
         publishDir       = [
-            enabled: params.generate_gvcf,
             mode: params.publish_dir_mode,
             path: { "${params.outdir}/variant_calling/${meta.id}/haplotypecaller"},
             pattern: "*{vcf.gz,vcf.gz.tbi}"

diff --git a/conf/test.config b/conf/test.config
@@ -115,10 +115,10 @@ profiles {
     tools_germline {
         params.input               = "${baseDir}/tests/csv/3.0/recalibrated_germline.csv"
         params.dbsnp               = params.test_data['homo_sapiens']['genome']['dbsnp_138_hg38_21_vcf_gz']
+        params.known_indels        = params.test_data['homo_sapiens']['genome']['mills_and_1000g_indels_21_vcf_gz']
         params.fasta               = params.test_data['homo_sapiens']['genome']['genome_21_fasta']
         params.intervals           = params.test_data['homo_sapiens']['genome']['genome_21_multi_interval_bed']
         params.step                = 'variant_calling'
-        params.joint_germline      = true
         params.wes                 = true
 
         params.nucleotides_per_second = 20

diff --git a/modules.json b/modules.json
@@ -102,6 +102,9 @@
             "gatk4/calculatecontamination": {
                 "git_sha": "169b2b96c1167f89ab07127b7057c1d90a6996c7"
             },
+            "gatk4/cnnscorevariants": {
+                "git_sha": "5d72500d601432f5396e9022c3a709854197db1a"
+            },
             "gatk4/createsequencedictionary": {
                 "git_sha": "169b2b96c1167f89ab07127b7057c1d90a6996c7"
             },
@@ -111,6 +114,9 @@
             "gatk4/filtermutectcalls": {
                 "git_sha": "169b2b96c1167f89ab07127b7057c1d90a6996c7"
             },
+            "gatk4/filtervarianttranches": {
+                "git_sha": "175ea9b7f95a0e2fd3679f7a052c6dcb60b61a6e"
+            },
             "gatk4/gatherbqsrreports": {
                 "git_sha": "169b2b96c1167f89ab07127b7057c1d90a6996c7"
             },

diff --git a/modules/nf-core/modules/gatk4/cnnscorevariants/main.nf b/modules/nf-core/modules/gatk4/cnnscorevariants/main.nf
diff --git a/modules/nf-core/modules/gatk4/cnnscorevariants/meta.yml b/modules/nf-core/modules/gatk4/cnnscorevariants/meta.yml
diff --git a/modules/nf-core/modules/gatk4/filtervarianttranches/main.nf b/modules/nf-core/modules/gatk4/filtervarianttranches/main.nf
diff --git a/modules/nf-core/modules/gatk4/filtervarianttranches/meta.yml b/modules/nf-core/modules/gatk4/filtervarianttranches/meta.yml
diff --git a/nextflow.config b/nextflow.config
@@ -59,12 +59,12 @@ params {
     cf_minqual = 0                      // ControlFreec default values
     cf_window = null                    // by default we are not using this in Control-FREEC
 
-    generate_gvcf = false // g.vcf are not produced by HaplotypeCaller by default
     pon = null // No default PON (Panel of Normals) file for GATK Mutect2 / Sentieon TNscope
     pon_tbi = null // No default PON index for GATK Mutect2 / Sentieon TNscope
     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 //
+    joint_germline = false // g.vcf & joint germline calling are not run by default if HaplotypeCaller is selected
+
 
     // Annotation
     vep_dbnsfp = null // dbnsfp plugin disabled within VEP