Merge pull request #158 from CenterForMedicalGeneticsGhent/filter-imp…

…rovements Some additional filter improvements
nf-cmgg · Dec 19, 2023 · 146aef0 · 146aef0
2 parents 236955f + 5eadff3
commit 146aef0
Show file tree

Hide file tree

Showing 11 changed files with 37 additions and 47 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,14 +7,17 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### New Features
 
-1. Added the `--output_suffix` option to add a custom suffix to the basename of the output files.
+1. Added the `--output_suffix` parameter to add a custom suffix to the basename of the output files.
 2. Implemented files for the alphamissense plugin of VEP.
+3. Added the `--only_pass` parameter to only output variants that have the `PASS` flag in the FILTER column. (This is only applied when `--filter` is also given)
+4. Added the `--keep_alt_contigs` parameter. This will tell the pipeline to not filter out the alternate contigs, which will now be done by default.
 
 ### Improvements
 
 1. Updated the seqplorer profile so that the output filenames are correct for easy import
 2. Changed the separator in `--vcfanno_resources` to `;`
    instead of `,` to allow commas in glob patterns.
+3. Removed the reheader step from the vardict subworkflow and added a simple sed substitution to the vardictjava module
 
 ## v1.4.0 - Kingly Kortrijk - [December 6 2023]
 

diff --git a/assets/vardict.header.vcf.gz b/assets/vardict.header.vcf.gz
diff --git a/conf/hypercap.config b/conf/hypercap.config
@@ -7,6 +7,7 @@
 params {
     callers             = "vardict"
     scatter_count       = 5
+    only_pass           = true
 
     output_suffix       = "-vardict-decomposed-annotated"
 }
diff --git a/conf/modules.config b/conf/modules.config
@@ -109,7 +109,8 @@ process {
 
     withName: "^.*:CRAM_PREPARE_SAMTOOLS_BEDTOOLS:FILTER_BEDS\$" {
         ext.prefix = { "${meta.id}.filter"}
-        ext.args = "-d 150"
+        ext.args = "-vE \"NO_COVERAGE${params.keep_alt_contigs ? "" : "|alt|random|decoy|Un"}\""
+        ext.args2 = "-d 150"
         publishDir  = [
             overwrite: true,
             enabled: true,
@@ -267,12 +268,12 @@ process {
                 if(params.filter){
                     withName: "^.*:CRAM_CALL_GENOTYPE_GATK4:VCF_FILTER_BCFTOOLS:FILTER_1\$" {
                         ext.prefix = { "${meta.id}_filtered_snps" }
-                        ext.args = {'--output-type z --soft-filter \'GATKCutoffSNP\' -e \'TYPE="snp" && (MQRankSum < -12.5 || ReadPosRankSum < -8.0 || QD < 2.0 || FS > 60.0 || (QD < 10.0 && AD[0:1] / (AD[0:1] + AD[0:0]) < 0.25 && ReadPosRankSum < 0.0) || MQ < 30.0)\' -m \'+\''}
+                        ext.args = {"--output-type z --soft-filter 'GATKCutoffSNP' -e 'TYPE=\"snp\" && (MQRankSum < -12.5 || ReadPosRankSum < -8.0 || QD < 2.0 || FS > 60.0 || MQ < 30.0)' -m '+'"}
                     }
 
                     withName: "^.*:CRAM_CALL_GENOTYPE_GATK4:VCF_FILTER_BCFTOOLS:FILTER_2\$" {
                         ext.prefix = final_prefix
-                        ext.args = {'--output-type z --soft-filter \'GATKCutoffIndel\' -e \'TYPE="indel" && (ReadPosRankSum < -20.0 || QD < 2.0 || FS > 200.0 || SOR > 10.0 || (QD < 10.0 && AD[0:1] / (AD[0:1] + AD[0:0]) < 0.25 && ReadPosRankSum < 0.0))\' -m \'+\''}
+                        ext.args = {'--output-type z --soft-filter \'GATKCutoffIndel\' -e \'TYPE="indel" && (ReadPosRankSum < -20.0 || QD < 2.0 || FS > 200.0 || SOR > 10.0 )\' -m \'+\''}
                         publishDir = [
                             enabled: enableOutput("filter"),
                             overwrite: true,
@@ -312,23 +313,19 @@ process {
 
         withName: "^.*:CRAM_CALL_VARDICTJAVA:VCF_CONCAT_BCFTOOLS:BCFTOOLS_CONCAT\$" {
             ext.args   = '--allow-overlaps --output-type z'
-        }
-
-        withName: "^.*:CRAM_CALL_VARDICTJAVA:BCFTOOLS_REHEADER\$" {
+            ext.prefix = { "${meta.id}.${meta.caller}" }
             publishDir  = [
                 overwrite: true,
                 enabled: enableOutput("original"),
                 mode: params.publish_dir_mode,
                 path: final_output,
                 saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
             ] // SAVE
-            ext.prefix = { "${meta.id}.${meta.caller}" }
-            ext.args2  = '--output-type z'
         }
 
         if(params.filter) {
             withName: "^.*:CRAM_CALL_VARDICTJAVA:VCF_FILTER_BCFTOOLS:FILTER_1\$" {
-                ext.args = '-i \'QUAL >= 0\' --output-type z'
+                ext.args = "-i 'QUAL >= 0${params.only_pass ? " && FILTER=\"PASS\"" : ""}' --output-type z"
             }
 
             withName: "^.*:CRAM_CALL_VARDICTJAVA:VCF_FILTER_BCFTOOLS:FILTER_2\$" {

diff --git a/modules/local/filter_beds/main.nf b/modules/local/filter_beds/main.nf
@@ -20,11 +20,12 @@ process FILTER_BEDS {
     script:
     // Remove regions with no coverage from the callable regions BED file
     def args = task.ext.args ?: ''
+    def args2 = task.ext.args2 ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
 
     def unzip = bed.extension == "gz" ? "zcat" : "cat"
     """
-    ${unzip} ${bed} | grep -v NO_COVERAGE | bedtools merge ${args} > ${prefix}.bed
+    ${unzip} ${bed} | grep ${args} | bedtools merge ${args2} > ${prefix}.bed
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

diff --git a/modules/nf-core/vardictjava/main.nf b/modules/nf-core/vardictjava/main.nf
diff --git a/modules/nf-core/vardictjava/vardictjava.diff b/modules/nf-core/vardictjava/vardictjava.diff
diff --git a/nextflow.config b/nextflow.config
@@ -28,6 +28,8 @@ params {
     vardict_min_af             = 0.1   // Minimum allele frequency for VarDict
     normalize                  = false
     output_suffix              = ""
+    only_pass                  = false
+    keep_alt_contigs           = false
 
     // Module specific parameters
     dragstr                    = false

diff --git a/nextflow_schema.json b/nextflow_schema.json
@@ -238,14 +238,21 @@
                 },
                 "normalize": {
                     "type": "boolean",
-                    "description": "Normalize the VCF after joint genotyping (will run on the decomposed VCF when --decompose is also used)"
+                    "description": "Normalize the VCFs"
                 },
                 "output_suffix": {
                     "type": "string",
                     "description": "A custom suffix to add to the basename of the output files"
+                },
+                "only_pass": {
+                    "type": "boolean",
+                    "description": "Filter out all variants that don't have the PASS filter for vardict. This only works when --filter is also given"
+                },
+                "keep_alt_contigs": {
+                    "type": "boolean",
+                    "description": "Keep all aditional contigs for calling instead of filtering them out before"
                 }
-            },
-            "required": ["scatter_count"]
+            }
         },
         "institutional_config_options": {
             "title": "Institutional config options",

diff --git a/subworkflows/local/cram_call_vardictjava/main.nf b/subworkflows/local/cram_call_vardictjava/main.nf
@@ -73,28 +73,15 @@ workflow CRAM_CALL_VARDICTJAVA {
         )
         ch_versions = ch_versions.mix(VCF_CONCAT_BCFTOOLS.out.versions)
 
-        VCF_CONCAT_BCFTOOLS.out.vcfs
-            .combine(["${projectDir}/assets/vardict.header.vcf.gz"])
-            .map { meta, vcf, header ->
-                [ meta, vcf, header, [] ]
-            }
-            .set { ch_reheader_input}
-
-        BCFTOOLS_REHEADER(
-            ch_reheader_input,
-            ch_fai
-        )
-        ch_versions = ch_versions.mix(BCFTOOLS_REHEADER.out.versions.first())
-
         if(params.filter) {
             VCF_FILTER_BCFTOOLS(
-                BCFTOOLS_REHEADER.out.vcf,
+                VCF_CONCAT_BCFTOOLS.out.vcfs,
                 false
             )
             ch_versions = ch_versions.mix(VCF_FILTER_BCFTOOLS.out.versions)
             ch_filter_output = VCF_FILTER_BCFTOOLS.out.vcfs
         } else {
-            ch_filter_output = BCFTOOLS_REHEADER.out.vcf
+            ch_filter_output = VCF_CONCAT_BCFTOOLS.out.vcfs
         }
 
         TABIX_TABIX(

diff --git a/tests/fails.nf.test b/tests/fails.nf.test
@@ -8,14 +8,14 @@ nextflow_pipeline {
 
         when {
             params {
-                scatter_count = null
+                fasta = null
                 outdir = "${outputDir}"
             }
         }
 
         then {
             assert workflow.failed
-            assert workflow.stderr.join("\n").contains("* Missing required parameter: --scatter_count")
+            assert workflow.stderr.join("\n").contains("* Missing required parameter: --fasta")
         }
 
     }