Merge pull request nf-core#1325 from nf-core/config_update

Update config files and other minor code polishing
FriederikeHanssen · Nov 10, 2023 · 51c7676 · 51c7676
2 parents 611b8b1 + 055818f
commit 51c7676
Show file tree

Hide file tree

Showing 57 changed files with 650 additions and 629 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -26,6 +26,8 @@ Pårtetjåkko is a mountain in the south of the park.
 - [#1304](https://github.com/nf-core/sarek/pull/1304) - Update modules
 - [#1311](https://github.com/nf-core/sarek/pull/1311) - Update local modules with an `environment.yml` file
 - [#1317](https://github.com/nf-core/sarek/pull/1317) - Add new tools to subway map
+- [#1325](https://github.com/nf-core/sarek/pull/1325) - Move `sentieon_dnascope_model` params into `igenomes.config`
+- [#1325](https://github.com/nf-core/sarek/pull/1325) - Refactor config files
 
 ### Fixed
 
@@ -36,6 +38,7 @@ Pårtetjåkko is a mountain in the south of the park.
 - [#1315](https://github.com/nf-core/sarek/pull/1315) - Avoid clash of configs of `FILTERVARIANTTRANCHES` in the Sentieon-Haplotyper and GATK-Haplotypecaller subworkflows
 - [#1318](https://github.com/nf-core/sarek/pull/1218) - Fix writing of params.json on S3
 - [#1324](https://github.com/nf-core/sarek/pull/1324) - Fix various typos & code formatting
+- [#1325](https://github.com/nf-core/sarek/pull/1325) - Update bcfannotate tests and related config files
 
 ### Removed
 

diff --git a/conf/base.config b/conf/base.config
@@ -21,84 +21,84 @@ process {
 
     // Process-specific resource requirements
     // See https://www.nextflow.io/docs/latest/config.html#config-process-selectors
+    withLabel:error_ignore {
+        errorStrategy = 'ignore'
+    }
+    withLabel:error_retry {
+        errorStrategy = 'retry'
+        maxRetries    = 2
+    }
     withLabel:process_single {
-        cpus          = { check_max( 1                  , 'cpus'    ) }
-        memory        = { check_max( 6.GB * task.attempt, 'memory'  ) }
-        time          = { check_max( 4.h  * task.attempt, 'time'    ) }
+        cpus   = { check_max( 1                  , 'cpus'    ) }
+        memory = { check_max( 6.GB * task.attempt, 'memory'  ) }
+        time   = { check_max( 4.h  * task.attempt, 'time'    ) }
     }
     withLabel:process_low {
-        cpus          = { check_max( 2     * task.attempt, 'cpus'    ) }
-        memory        = { check_max( 12.GB * task.attempt, 'memory'  ) }
-        time          = { check_max( 4.h   * task.attempt, 'time'    ) }
+        cpus   = { check_max( 2     * task.attempt, 'cpus'    ) }
+        memory = { check_max( 12.GB * task.attempt, 'memory'  ) }
+        time   = { check_max( 4.h   * task.attempt, 'time'    ) }
     }
     withLabel:process_medium {
-        cpus          = { check_max( 6     * task.attempt, 'cpus'    ) }
-        memory        = { check_max( 36.GB * task.attempt, 'memory'  ) }
-        time          = { check_max( 8.h   * task.attempt, 'time'    ) }
+        cpus   = { check_max( 6     * task.attempt, 'cpus'    ) }
+        memory = { check_max( 36.GB * task.attempt, 'memory'  ) }
+        time   = { check_max( 8.h   * task.attempt, 'time'    ) }
     }
     withLabel:process_high {
-        cpus          = { check_max( 12    * task.attempt, 'cpus'    ) }
-        memory        = { check_max( 72.GB * task.attempt, 'memory'  ) }
-        time          = { check_max( 16.h  * task.attempt, 'time'    ) }
+        cpus   = { check_max( 12    * task.attempt, 'cpus'    ) }
+        memory = { check_max( 72.GB * task.attempt, 'memory'  ) }
+        time   = { check_max( 16.h  * task.attempt, 'time'    ) }
     }
     withLabel:process_long {
-        time          = { check_max( 20.h  * task.attempt, 'time'    ) }
+        time   = { check_max( 20.h  * task.attempt, 'time'    ) }
     }
     withLabel:process_high_memory {
-        memory        = { check_max( 200.GB * task.attempt, 'memory' ) }
-    }
-    withLabel:error_ignore {
-        errorStrategy = 'ignore'
-    }
-    withLabel:error_retry {
-        errorStrategy = 'retry'
-        maxRetries    = 2
+        memory = { check_max( 200.GB * task.attempt, 'memory' ) }
     }
     withName: 'UNZIP.*|UNTAR.*|TABIX.*|BUILD_INTERVALS|CREATE_INTERVALS_BED|CUSTOM_DUMPSOFTWAREVERSIONS|VCFTOOLS|BCFTOOLS.*|SAMTOOLS_INDEX' {
-        cpus            = { check_max( 1 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 1.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 1 * task.attempt, 'cpus' ) }
+        memory = { check_max( 1.GB * task.attempt, 'memory' ) }
     }
     withName: 'FASTQC'{
-        cpus            = { check_max( 4 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 4 * task.attempt, 'cpus' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName: 'FASTP'{
-        cpus            = { check_max( 12 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 12 * task.attempt, 'cpus' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName: 'BWAMEM1_MEM|BWAMEM2_MEM' {
-        cpus            = { check_max( 24 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 30.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 24 * task.attempt, 'cpus' ) }
+        memory = { check_max( 30.GB * task.attempt, 'memory' ) }
     }
     withName:'CNVKIT_BATCH' {
-        label           = "process_high"
-        memory          = { check_max( 36.GB * task.attempt, 'memory' ) }
+        label  = "process_high"
+        memory = { check_max( 36.GB * task.attempt, 'memory' ) }
     }
     withName: 'GATK4_MARKDUPLICATES|GATK4SPARK_MARKDUPLICATES' {
-        cpus           = { check_max( 6 * task.attempt, 'cpus' ) }
-        memory         = { check_max( 30.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 6 * task.attempt, 'cpus' ) }
+        memory = { check_max( 30.GB * task.attempt, 'memory' ) }
     }
     withName:'GATK4_APPLYBQSR|GATK4SPARK_APPLYBQSR|GATK4_BASERECALIBRATOR|GATK4SPARK_BASERECALIBRATOR|GATK4_GATHERBQSRREPORTS'{
-        cpus            = { check_max( 2 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 2 * task.attempt, 'cpus' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName:'MOSDEPTH'{
-        cpus            = { check_max( 4 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 4 * task.attempt, 'cpus' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName:'STRELKA.*|MANTA.*' {
-        cpus            = { check_max( 10 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 8.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 10 * task.attempt, 'cpus' ) }
+        memory = { check_max( 8.GB * task.attempt, 'memory' ) }
     }
     withName:'SAMTOOLS_CONVERT'{
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName:'GATK4_MERGEVCFS'{
-        cpus            = { check_max( 2 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 4.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 2 * task.attempt, 'cpus' ) }
+        memory = { check_max( 4.GB * task.attempt, 'memory' ) }
     }
     withName: 'MULTIQC' {
-        cpus            = { check_max( 4 * task.attempt, 'cpus' ) }
-        memory          = { check_max( 12.GB * task.attempt, 'memory' ) }
+        cpus   = { check_max( 4 * task.attempt, 'cpus' ) }
+        memory = { check_max( 12.GB * task.attempt, 'memory' ) }
     }
 }
diff --git a/conf/igenomes.config b/conf/igenomes.config
@@ -43,40 +43,41 @@ params {
             vep_species           = 'homo_sapiens'
         }
         'GATK.GRCh38' {
-            ascat_alleles         = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/G1000_alleles_hg38.zip"
-            ascat_genome          = 'hg38'
-            ascat_loci            = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/G1000_loci_hg38.zip"
-            ascat_loci_gc         = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/GC_G1000_hg38.zip"
-            ascat_loci_rt         = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/RT_G1000_hg38.zip"
-            bwa                   = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/BWAIndex/"
-            bwamem2               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/BWAmem2Index/"
-            dragmap               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/dragmap/"
-            chr_dir               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/Chromosomes"
-            cf_chrom_len          = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/Length/Homo_sapiens_assembly38.len"
-            dbsnp                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/dbsnp_146.hg38.vcf.gz"
-            dbsnp_tbi             = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/dbsnp_146.hg38.vcf.gz.tbi"
-            dbsnp_vqsr            = '--resource:dbsnp,known=false,training=true,truth=false,prior=2.0 dbsnp_146.hg38.vcf.gz'
-            dict                  = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.dict"
-            fasta                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.fasta"
-            fasta_fai             = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.fasta.fai"
-            germline_resource     = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/af-only-gnomad.hg38.vcf.gz"
-            germline_resource_tbi = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/af-only-gnomad.hg38.vcf.gz.tbi"
-            intervals             = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/intervals/wgs_calling_regions_noseconds.hg38.bed"
-            known_snps            = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000G_omni2.5.hg38.vcf.gz"
-            known_snps_tbi        = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000G_omni2.5.hg38.vcf.gz.tbi"
-            known_snps_vqsr       = '--resource:1000G,known=false,training=true,truth=true,prior=10.0 1000G_omni2.5.hg38.vcf.gz'
-            known_indels          = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/{Mills_and_1000G_gold_standard.indels.hg38,beta/Homo_sapiens_assembly38.known_indels}.vcf.gz"
-            known_indels_tbi      = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/{Mills_and_1000G_gold_standard.indels.hg38,beta/Homo_sapiens_assembly38.known_indels}.vcf.gz.tbi"
-            known_indels_vqsr     = '--resource:gatk,known=false,training=true,truth=true,prior=10.0 Homo_sapiens_assembly38.known_indels.vcf.gz --resource:mills,known=false,training=true,truth=true,prior=10.0 Mills_and_1000G_gold_standard.indels.hg38.vcf.gz'
-            mappability           = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/Control-FREEC/out100m2_hg38.gem"
-            ngscheckmate_bed      = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/NGSCheckMate/SNP_GRCh38_hg38_wChr.bed"
-            pon                   = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000g_pon.hg38.vcf.gz"
-            pon_tbi               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000g_pon.hg38.vcf.gz.tbi"
-            snpeff_db             = 105
-            snpeff_genome         = 'GRCh38'
-            vep_cache_version     = 110
-            vep_genome            = 'GRCh38'
-            vep_species           = 'homo_sapiens'
+            ascat_alleles           = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/G1000_alleles_hg38.zip"
+            ascat_genome            = 'hg38'
+            ascat_loci              = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/G1000_loci_hg38.zip"
+            ascat_loci_gc           = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/GC_G1000_hg38.zip"
+            ascat_loci_rt           = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/ASCAT/RT_G1000_hg38.zip"
+            bwa                     = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/BWAIndex/"
+            bwamem2                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/BWAmem2Index/"
+            cf_chrom_len            = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/Length/Homo_sapiens_assembly38.len"
+            chr_dir                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/Chromosomes"
+            dbsnp                   = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/dbsnp_146.hg38.vcf.gz"
+            dbsnp_tbi               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/dbsnp_146.hg38.vcf.gz.tbi"
+            dbsnp_vqsr              = '--resource:dbsnp,known=false,training=true,truth=false,prior=2.0 dbsnp_146.hg38.vcf.gz'
+            dict                    = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.dict"
+            dragmap                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/dragmap/"
+            fasta                   = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.fasta"
+            fasta_fai               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Sequence/WholeGenomeFasta/Homo_sapiens_assembly38.fasta.fai"
+            germline_resource       = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/af-only-gnomad.hg38.vcf.gz"
+            germline_resource_tbi   = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/af-only-gnomad.hg38.vcf.gz.tbi"
+            intervals               = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/intervals/wgs_calling_regions_noseconds.hg38.bed"
+            known_indels            = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/{Mills_and_1000G_gold_standard.indels.hg38,beta/Homo_sapiens_assembly38.known_indels}.vcf.gz"
+            known_indels_tbi        = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/{Mills_and_1000G_gold_standard.indels.hg38,beta/Homo_sapiens_assembly38.known_indels}.vcf.gz.tbi"
+            known_indels_vqsr       = '--resource:gatk,known=false,training=true,truth=true,prior=10.0 Homo_sapiens_assembly38.known_indels.vcf.gz --resource:mills,known=false,training=true,truth=true,prior=10.0 Mills_and_1000G_gold_standard.indels.hg38.vcf.gz'
+            known_snps              = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000G_omni2.5.hg38.vcf.gz"
+            known_snps_tbi          = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000G_omni2.5.hg38.vcf.gz.tbi"
+            known_snps_vqsr         = '--resource:1000G,known=false,training=true,truth=true,prior=10.0 1000G_omni2.5.hg38.vcf.gz'
+            mappability             = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/Control-FREEC/out100m2_hg38.gem"
+            ngscheckmate_bed        = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/NGSCheckMate/SNP_GRCh38_hg38_wChr.bed"
+            pon                     = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000g_pon.hg38.vcf.gz"
+            pon_tbi                 = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/GATKBundle/1000g_pon.hg38.vcf.gz.tbi"
+            sentieon_dnascope_model = "${params.igenomes_base}/Homo_sapiens/GATK/GRCh38/Annotation/Sentieon/SentieonDNAscopeModel1.1.model"
+            snpeff_db               = 105
+            snpeff_genome           = 'GRCh38'
+            vep_cache_version       = 110
+            vep_genome              = 'GRCh38'
+            vep_species             = 'homo_sapiens'
         }
         'Ensembl.GRCh37' {
             bwa                   = "${params.igenomes_base}/Homo_sapiens/Ensembl/GRCh37/Sequence/BWAIndex/version0.6.0/"

diff --git a/conf/modules/aligner.config b/conf/modules/aligner.config
@@ -16,26 +16,26 @@
 process {
 
     if (params.step == 'mapping') {
-        withName: "BWAMEM1_MEM" {
-            ext.when         = { params.aligner == "bwa-mem" }
+        withName: 'BWAMEM1_MEM' {
+            ext.when   = { params.aligner == 'bwa-mem' }
         }
 
-        withName: "BWAMEM2_MEM" {
-            ext.when         = { params.aligner == "bwa-mem2" }
+        withName: 'BWAMEM2_MEM' {
+            ext.when   = { params.aligner == 'bwa-mem2' }
         }
 
-        withName: "DRAGMAP_ALIGN" {
-            ext.when         = { params.aligner == "dragmap" }
-            ext.args         = { "--RGSM ${meta.patient}_${meta.sample} --RGID ${meta.read_group}" }
+        withName: 'DRAGMAP_ALIGN' {
+            ext.args   = { "--RGSM ${meta.patient}_${meta.sample} --RGID ${meta.read_group}" }
+            ext.when   = { params.aligner == 'dragmap' }
         }
 
-        withName: "SENTIEON_BWAMEM" {
-            ext.when         = { params.aligner == "sentieon-bwamem" }
+        withName: 'SENTIEON_BWAMEM' {
+            ext.when   = { params.aligner == 'sentieon-bwamem' }
         }
 
-        withName: "(BWAMEM.*_MEM|DRAGMAP_ALIGN|SENTIEON_BWAMEM)" {
-            ext.prefix       = { params.split_fastq > 1 ? "${meta.id}".concat('.').concat(reads.get(0).name.tokenize('.')[0]) : "${meta.id}.sorted" }
-            publishDir       = [
+        withName: 'BWAMEM.*_MEM|DRAGMAP_ALIGN|SENTIEON_BWAMEM' {
+            ext.prefix = { params.split_fastq > 1 ? "${meta.id}".concat('.').concat(reads.get(0).name.tokenize('.')[0]) : "${meta.id}.sorted" }
+            publishDir = [
                 mode: params.publish_dir_mode,
                 path: { "${params.outdir}/preprocessing/" },
                 pattern: "*bam",
@@ -55,22 +55,22 @@ process {
             ]
         }
 
-        withName: "(BWAMEM.*_MEM|DRAGMAP_ALIGN)" {
+        withName: 'BWAMEM.*_MEM|DRAGMAP_ALIGN' {
             // Markduplicates Spark NEEDS name-sorted reads or runtime goes through the roof
             // However if it's skipped, reads need to be coordinate-sorted
             // Only name sort if Spark for Markduplicates + duplicate marking is not skipped
             // Currently SENTIEON_BWAMEM only supports coordinate sorting the reads.
-            ext.args2        = { params.use_gatk_spark && params.use_gatk_spark.contains('markduplicates') && (!params.skip_tools || (params.skip_tools && !params.skip_tools.split(',').contains('markduplicates'))) ? '-n' : '' }
+            ext.args2  = { params.use_gatk_spark && params.use_gatk_spark.contains('markduplicates') && (!params.skip_tools || (params.skip_tools && !params.skip_tools.split(',').contains('markduplicates'))) ? '-n' : '' }
         }
 
-        withName: "BWAMEM.*_MEM|SENTIEON_BWAMEM" {
+        withName: 'BWAMEM.*_MEM|SENTIEON_BWAMEM' {
             // Using -B 3 for tumor samples
-            ext.args         = { meta.status == 1 ? "-K 100000000 -Y -B 3 -R ${meta.read_group}" : "-K 100000000 -Y -R ${meta.read_group}" }
+            ext.args   = { meta.status == 1 ? "-K 100000000 -Y -B 3 -R ${meta.read_group}" : "-K 100000000 -Y -R ${meta.read_group}" }
         }
     }
 
     withName: 'MERGE_BAM|INDEX_MERGE_BAM' {
-        publishDir       = [
+        publishDir = [
             mode: params.publish_dir_mode,
             path: { "${params.outdir}/preprocessing/" },
             pattern: "*{bam,bai}",