Add skip_tools haplotypecaller_filter

nextflow_schema.json

@@ -96,7 +96,7 @@
                     "fa_icon": "fas fa-forward",
                     "description": "Disable specified tools.",
                     "help_text": "Multiple tools can be specified, separated by commas.\n\n> **NB** `--skip_tools baserecalibrator_report` is actually just not saving the reports.\n> **NB** `--skip_tools markduplicates_report` does not skip `MarkDuplicates` but prevent the collection of duplicate metrics that slows down performance.",
-                    "pattern": "^((baserecalibrator|baserecalibrator_report|bcftools|documentation|fastqc|markduplicates|markduplicates_report|mosdepth|multiqc|samtools|vcftools|versions)?,?)*[^,]+$"
+                    "pattern": "^((baserecalibrator|baserecalibrator_report|bcftools|documentation|fastqc|haplotypecaller_filter|markduplicates|markduplicates_report|mosdepth|multiqc|samtools|vcftools|versions)?,?)*[^,]+$"
                 }
             },
             "fa_icon": "fas fa-user-cog"

subworkflows/local/bam_variant_calling_germline_all/main.nf

@@ -14,6 +14,7 @@ include { BAM_VARIANT_CALLING_SINGLE_TIDDIT   } from '../bam_variant_calling_sin
 workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
     take:
         tools                             // Mandatory, list of tools to apply
+        skip_tools                        // Mandatory, list of tools to skip
         cram_recalibrated                 // channel: [mandatory] cram
         bwa                               // channel: [mandatory] bwa
         dbsnp                             // channel: [mandatory] dbsnp
@@ -180,7 +181,8 @@ workflow BAM_VARIANT_CALLING_GERMLINE_ALL {
                         known_sites_indels_tbi,
                         known_sites_snps,
                         known_sites_snps_tbi,
-                        intervals_bed_combined_haplotypec)
+                        intervals_bed_combined_haplotypec,
+                        skip_tools)
 
         haplotypecaller_vcf      = BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.filtered_vcf
         ch_versions              = ch_versions.mix(BAM_VARIANT_CALLING_HAPLOTYPECALLER.out.versions)

subworkflows/local/bam_variant_calling_haplotypecaller/main.nf

@@ -17,7 +17,7 @@ workflow BAM_VARIANT_CALLING_HAPLOTYPECALLER {
     known_sites_snps
     known_sites_snps_tbi
     intervals_bed_combined          // channel: [mandatory] intervals/target regions in one file unzipped, no_intervals.bed if no_intervals
-
+    skip_tools
 
     main:
 
@@ -126,31 +126,34 @@ workflow BAM_VARIANT_CALLING_HAPLOTYPECALLER {
 
         realigned_bam = BAM_MERGE_INDEX_SAMTOOLS.out.bam_bai
 
-        VCF_VARIANT_FILTERING_GATK(haplotypecaller_vcf.join(haplotypecaller_tbi),
-                    fasta,
-                    fasta_fai,
-                    dict,
-                    intervals_bed_combined,
-                    known_sites_indels.concat(known_sites_snps).flatten().unique().collect(),
-                    known_sites_indels_tbi.concat(known_sites_snps_tbi).flatten().unique().collect())
-
-        filtered_vcf = VCF_VARIANT_FILTERING_GATK.out.filtered_vcf.map{ meta, vcf-> [
-                [
-                    patient:meta.patient,
-                    sample:meta.sample,
-                    status:meta.status,
-                    sex:meta.sex,
-                    id:meta.sample,
-                    num_intervals:meta.num_intervals,
-                    variantcaller:"haplotypecaller"
-                ],
-                vcf
-            ]
+        if (!(skip_tools && skip_tools.split(',').contains('haplotypecaller_filter'))) {
+
+            VCF_VARIANT_FILTERING_GATK(haplotypecaller_vcf.join(haplotypecaller_tbi),
+                        fasta,
+                        fasta_fai,
+                        dict,
+                        intervals_bed_combined,
+                        known_sites_indels.concat(known_sites_snps).flatten().unique().collect(),
+                        known_sites_indels_tbi.concat(known_sites_snps_tbi).flatten().unique().collect())
+
+            filtered_vcf = VCF_VARIANT_FILTERING_GATK.out.filtered_vcf.map{ meta, vcf-> [
+                    [
+                        patient:meta.patient,
+                        sample:meta.sample,
+                        status:meta.status,
+                        sex:meta.sex,
+                        id:meta.sample,
+                        num_intervals:meta.num_intervals,
+                        variantcaller:"haplotypecaller"
+                    ],
+                    vcf
+                ]
+            }
+            ch_versions = ch_versions.mix(VCF_VARIANT_FILTERING_GATK.out.versions)
         }
 
         ch_versions = ch_versions.mix(GATK4_HAPLOTYPECALLER.out.versions)
         ch_versions = ch_versions.mix(MERGE_HAPLOTYPECALLER.out.versions)
-        ch_versions = ch_versions.mix(VCF_VARIANT_FILTERING_GATK.out.versions)
     }
 
     emit:

tests/test_haplotypecaller.yml

@@ -34,6 +34,34 @@
     # binary changes md5sums on reruns.
     - path: results/haplotypecaller
       should_exist: false
+- name: Run variant calling on germline sample with haplotypecaller and skip filter
+  command: nextflow run main.nf -profile test,targeted --input ./tests/csv/3.0/mapped_single_bam.csv --tools haplotypecaller --step variant_calling --skip_tools haplotypecaller_filter
+  tags:
+    - germline
+    - haplotypecaller
+    - variant_calling
+  files:
+    - path: results/csv/variantcalled.csv
+      md5sum: d7d86e82902a4f57876b2414a4f812a4
+    - path: results/multiqc
+    - path: results/preprocessing/converted/test/test.converted.cram
+    # binary changes md5sums on reruns.
+    - path: results/preprocessing/converted/test/test.converted.cram.crai
+    # binary changes md5sums on reruns.
+    - path: results/preprocessing/recalibrated/test/test.recal.cram
+      should_exist: false
+    - path: results/preprocessing/recalibrated/test/test.recal.cram.crai
+      should_exist: false
+    - path: results/variant_calling/haplotypecaller/test/test.haplotypecaller.filtered.vcf.gz
+      should_exist: false
+    - path: results/variant_calling/haplotypecaller/test/test.haplotypecaller.filtered.vcf.gz.tbi
+      should_exist: false
+    - path: results/variant_calling/haplotypecaller/test/test.haplotypecaller.vcf.gz
+    # binary changes md5sums on reruns.
+    - path: results/variant_calling/haplotypecaller/test/test.haplotypecaller.vcf.gz.tbi
+    # binary changes md5sums on reruns.
+    - path: results/haplotypecaller
+      should_exist: false
 - name: Run variant calling on germline sample with haplotypecaller without intervals
   command: nextflow run main.nf -profile test,targeted --input ./tests/csv/3.0/mapped_single_bam.csv --tools haplotypecaller --step variant_calling --no_intervals
   tags:

workflows/sarek.nf

@@ -954,6 +954,7 @@ workflow SAREK {
         // GERMLINE VARIANT CALLING
         BAM_VARIANT_CALLING_GERMLINE_ALL(
             params.tools,
+            params.skip_tools,
             ch_cram_variant_calling_status_normal,
             [[id:"bwa"],[]], //bwa_index for tiddit; not used here
             dbsnp,