Merge pull request #88 from CenterForMedicalGeneticsGhent/dev

Release v1.1.1

CHANGELOG.md

@@ -3,6 +3,18 @@
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## v1.1.1 - Golden Ghent - [Mar 20 2023]
+
+### Changes
+
+1. Set the default of `--validate` to `false`
+
+### Bug fixes
+
+1. Fixed a bug with ensembl VEP. Filenames of the alt contigs should now have a `_alt` suffix instead of all alt contigs.
+2. Added file-exist check to the `sdf` file
+3. Fixed the scattering when using alt contigs
+
 ## v1.1.0 - Glorious Ghent - [Mar 14 2023]
 
 ### New Features

conf/modules.config

@@ -34,7 +34,7 @@ process {
     //     ext.args = "-d ${params.bed_merge_distance}"
     // }
 
-    withName: GOLEFT_INDEXSPLIT {
+    withName: "^.*:PREPROCESSING:GOLEFT_INDEXSPLIT\$" {
         publishDir  = [
             overwrite: true,
             enabled: true,
@@ -58,8 +58,7 @@ process {
                 '-ERC GVCF -contamination "0"',
                 '-GQB 10 -GQB 20 -GQB 30 -GQB 40 -GQB 50 -GQB 60 -GQB 70 -GQB 80 -GQB 90',
                 '-G StandardAnnotation -G StandardHCAnnotation -G AS_StandardAnnotation',
-                '--dragen-mode',
-                "--intervals ${meta.region}"
+                '--dragen-mode'
             ].join(" ")
         }
     }
@@ -134,9 +133,7 @@ process {
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
     */
 
-    withName: "^.*JOINT_GENOTYPING:BEDTOOLS_MAKEWINDOWS\$" {
-        ext.prefix = { "${meta.id}.windows" }
-        ext.args = "-n ${params.scatter_count}"
+    withName: "^.*:JOINT_GENOTYPING:GOLEFT_INDEXSPLIT\$" {
         publishDir = [
             enabled: true,
             overwrite: true,
@@ -165,7 +162,6 @@ process {
                 "--allow-old-rms-mapping-quality-annotation-data",
                 "-G StandardAnnotation -G AS_StandardAnnotation",
                 "-AX ExcessHet -AX InbreedingCoeff",
-                "--intervals ${meta.region}",
                 "--interval-padding 20"
             ].join(" ")
         }
@@ -287,7 +283,7 @@ process {
             // maxRetries = 3
             // container = {params.vep_cache ? "quay.io/biocontainers/ensembl-vep:${params.vep_version}--pl5321h4a94de4_${task.attempt - 1}" : "nfcore/vep:${params.vep_version}.${params.genome}"}
             container = "nfcore/vep:${params.vep_version}.${params.genome}"
-            ext.prefix = {"${meta.id}_${meta.regions}"}
+            ext.prefix = {"${meta.id}_${meta.regions.contains(",") ? "alt" : meta.regions }"}
             ext.args = {[
                 // select the chromosomes to run on
                 "--chr ${meta.regions}",

modules.json

@@ -36,11 +36,6 @@
                         "installed_by": ["modules"],
                         "patch": "modules/nf-core/bedtools/intersect/bedtools-intersect.diff"
                     },
-                    "bedtools/makewindows": {
-                        "branch": "master",
-                        "git_sha": "007c99a9726dfd89198e788162db594cd29d426f",
-                        "installed_by": ["modules"]
-                    },
                     "bedtools/merge": {
                         "branch": "master",
                         "git_sha": "c8e35eb2055c099720a75538d1b8adb3fb5a464c",

modules/local/split_beds/main.nf

@@ -0,0 +1,44 @@
+process SPLIT_BEDS {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "anaconda::gawk=5.1.0"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gawk:5.1.0' :
+        'quay.io/biocontainers/gawk:5.1.0' }"
+
+    input:
+    tuple val(meta), path(bed), val(sample_count)
+
+    output:
+    tuple val(meta), path("*.bed")  , emit: beds
+    path('versions.yml')            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def prefix = task.ext.prefix ?: meta.id
+    def threshold = sample_count * 0.3
+    // This module will split a BED file created with goleft/indexsplit into 
+    // multiple BED files. All regions that have a scaled data size lower than 0.3 * amount of samples
+    // will be merged into one BED file. All regions that have a higher scaled data size
+    // will be put into their own BED file. Also all regions with no reads are removed.
+    """
+    awk -vFS="\t" '{
+        if (\$0 ~ /^[^#].*\$/ && \$5 >= 1) {
+            if (\$4 >= ${threshold}) {
+                print \$0 > sprintf("${prefix}_%s_%s_%s.bed", \$1, \$2, \$3)
+            } else {
+                print \$0 > "${prefix}_others.bed"
+            }
+        }
+    }' ${bed}
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gawk: \$(awk -Wversion | sed '1!d; s/.*Awk //; s/,.*//')
+    END_VERSIONS
+    """
+
+    //No stub because this process is so small (would be harder to stub than to just let it run)
+}

nextflow.config

@@ -19,7 +19,7 @@ params {
     annotate                   = false
     gemini                     = false
     add_ped                    = false
-    validate                   = true
+    validate                   = false
     roi                        = null
 
     // Module specific parameters
@@ -244,7 +244,7 @@ manifest {
     description     = 'A nextflow pipeline for calling and annotating variants'
     mainScript      = 'main.nf'
     nextflowVersion = '!>=22.10.5'
-    version = '1.1.0'
+    version = '1.1.1'
 }
 
 // Load modules.config for DSL2 module specific options

nextflow_schema.json

@@ -123,7 +123,8 @@
                     "hidden": true,
                     "help_text": "Do not load `igenomes.config` when running the pipeline. You may choose this option if you observe clashes between custom parameters and those supplied in `igenomes.config`."
                 }
-            }
+            },
+            "required": ["fasta"]
         },
         "pipeline_specific_parameters": {
             "title": "Pipeline specific parameters",
@@ -134,7 +135,7 @@
                 "scatter_count": {
                     "type": "integer",
                     "default": 40,
-                    "description": "The amount of scattering that should happen per sample. Increase this number to increase the pipeline run speed, but at the tradeoff of using more IO and disk space"
+                    "description": "The amount of scattering that should happen per sample. Increase this number to increase the pipeline run speed, but at the tradeoff of using more IO and disk space. The actual scatter value can differ a bit (usually not a bigger difference than 1) when using alt contigs in the reference."
                 },
                 "species": {
                     "type": "string",
@@ -180,7 +181,8 @@
                 "use_dragstr_model": {
                     "type": "boolean",
                     "description": "Whether or not the DRAGSTR models should be used",
-                    "fa_icon": "fas fa-question-circle"
+                    "fa_icon": "fas fa-question-circle",
+                    "default": true
                 },
                 "dbsnp": {
                     "type": "string",

subworkflows/local/germline_variant_calling.nf

@@ -1,6 +1,7 @@
 //
 // GERMLINE VARIANT CALLING
 //
+include { SPLIT_BEDS                                            } from '../../modules/local/split_beds/main'
 
 include { GATK4_HAPLOTYPECALLER as HAPLOTYPECALLER              } from '../../modules/nf-core/gatk4/haplotypecaller/main'
 include { GATK4_CALIBRATEDRAGSTRMODEL as CALIBRATEDRAGSTRMODEL  } from '../../modules/nf-core/gatk4/calibratedragstrmodel/main'
@@ -26,6 +27,15 @@ workflow GERMLINE_VARIANT_CALLING {
     ch_versions  = Channel.empty()
     ch_reports   = Channel.empty()
 
+    //
+    // Split BED file into correct regions
+    //
+
+    SPLIT_BEDS(
+        beds.map { it + [1] }
+    )
+    ch_versions = ch_versions.mix(SPLIT_BEDS.out.versions.first())
+
     //
     // Generate DRAGSTR models
     //
@@ -47,13 +57,13 @@ workflow GERMLINE_VARIANT_CALLING {
         ch_versions = ch_versions.mix(CALIBRATEDRAGSTRMODEL.out.versions)
 
         crams
-            .join(beds, failOnDuplicate: true, failOnMismatch: true)
+            .join(SPLIT_BEDS.out.beds, failOnDuplicate: true, failOnMismatch: true)
             .join(CALIBRATEDRAGSTRMODEL.out.dragstr_model, failOnDuplicate: true, failOnMismatch: true)
             .set { cram_models }
     }
     else {
         crams
-            .join(beds, failOnDuplicate: true, failOnMismatch: true)
+            .join(SPLIT_BEDS.out.beds, failOnDuplicate: true, failOnMismatch: true)
             .set { cram_models }
     }
 
@@ -63,17 +73,16 @@ workflow GERMLINE_VARIANT_CALLING {
 
     cram_models
         .dump(tag:'cram_models', pretty:true)
-        .map { meta, cram, crai, bed, dragstr_model=[] ->
-            new_meta = meta + [region_count: bed.readLines().size()]
-            [ new_meta, cram, crai, bed, dragstr_model ]
+        .map { meta, cram, crai, beds, dragstr_model=[] ->
+            bed_is_list = beds instanceof ArrayList
+            new_meta = meta + [region_count: bed_is_list ? beds.size() : 1]
+            [ new_meta, cram, crai, bed_is_list ? beds : [beds], dragstr_model ]
         }
-        .splitText(elem:3)
+        .transpose(by:3)
         .map(
-            { meta, cram, crai, region, dragstr_model ->
-                region_split = region[0..-2].split("\t")
-                region_string = "${region_split[0]}:${region_split[1] as int + 1}-${region_split[2]}"
-                new_meta = meta + [id:"${meta.id}_${region_string.replace(":","_").replace("-":"_")}", region:region_string]
-                [ new_meta, cram, crai, [], dragstr_model ]
+            { meta, cram, crai, bed, dragstr_model ->
+                new_meta = meta + [id:bed.baseName]
+                [ new_meta, cram, crai, bed, dragstr_model ]
             }
         )
         .dump(tag:'cram_intervals', pretty:true)
@@ -159,4 +168,3 @@ workflow GERMLINE_VARIANT_CALLING {
     versions = ch_versions
     reports  = ch_reports
 }
-