Merge pull request #5 from TRON-Bioinformatics/upgrade-gatk

upgrade GATK to 4.2.5.0

Author: priesgo

main.nf

@@ -30,8 +30,8 @@ params.collect_hs_metrics_min_mapping_quality = false
 // computational resources
 params.prepare_bam_cpus = 3
 params.prepare_bam_memory = "8g"
-params.mark_duplicates_cpus = 16
-params.mark_duplicates_memory = "64g"
+params.mark_duplicates_cpus = 2
+params.mark_duplicates_memory = "16g"
 params.realignment_around_indels_cpus = 2
 params.realignment_around_indels_memory = "31g"
 params.bqsr_cpus = 3

modules/01_prepare_bam.nf

@@ -17,7 +17,7 @@ process PREPARE_BAM {
     memory "${params.prepare_bam_memory}"
     tag "${name}"
 
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
 
     input:
     tuple val(name), val(type), file(bam)
@@ -58,7 +58,7 @@ process INDEX_BAM {
     memory "${params.index_memory}"
     tag "${name}"
 
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
 
     input:
     tuple val(name), val(type), file(bam)

modules/02_mark_duplicates.nf

@@ -1,5 +1,5 @@
-params.mark_duplicates_cpus = 16
-params.mark_duplicates_memory = "64g"
+params.mark_duplicates_cpus = 2
+params.mark_duplicates_memory = "16g"
 params.remove_duplicates = true
 params.skip_metrics = false
 params.output = 'output'
@@ -11,7 +11,7 @@ process MARK_DUPLICATES {
     tag "${name}"
     publishDir "${params.output}/${name}/metrics/mark_duplicates", mode: "copy", pattern: "*.dedup_metrics.txt"
 
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
 
     input:
     tuple val(name), val(type), file(bam)
@@ -21,15 +21,25 @@ process MARK_DUPLICATES {
     file("${name}.dedup_metrics.txt") optional true
 
     script:
-    dedup_metrics = params.skip_metrics ? "": "--metrics-file ${name}.dedup_metrics.txt"
-    remove_duplicates = params.remove_duplicates ? "--remove-all-duplicates true" : "--remove-all-duplicates false"
+    dedup_metrics = params.skip_metrics ? "": "--METRICS_FILE ${name}.dedup_metrics.txt"
+    remove_duplicates = params.remove_duplicates ? "--REMOVE_DUPLICATES true" : "--REMOVE_DUPLICATES false"
     """
     mkdir tmp
 
-    gatk MarkDuplicatesSpark \
+    gatk SortSam \
+    --INPUT ${bam} \
+    --OUTPUT ${name}.sorted.bam \
+    --SORT_ORDER coordinate
+
+    gatk MarkDuplicates \
     --java-options '-Xmx${params.mark_duplicates_memory}  -Djava.io.tmpdir=tmp' \
-    --input  ${bam} \
-    --output ${name}.dedup.bam \
-    --conf 'spark.executor.cores=${task.cpus}' ${remove_duplicates} ${dedup_metrics}
+    --INPUT  ${name}.sorted.bam \
+    --OUTPUT ${name}.dedup.bam \
+    --ASSUME_SORT_ORDER coordinate \
+    --CREATE_INDEX true ${remove_duplicates} ${dedup_metrics}
+
+    cp ${name}.dedup.bai ${name}.dedup.bam.bai
+
+    rm -f ${name}.sorted.bam
     """
 }

modules/03_metrics.nf

@@ -13,7 +13,7 @@ process HS_METRICS {
     tag "${name}"
     publishDir "${params.output}/${name}/metrics/hs_metrics", mode: "copy"
 
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
 
     input:
     tuple val(name), val(type), file(bam), file(bai)
@@ -53,7 +53,7 @@ process METRICS {
     publishDir "${params.output}/${name}/metrics/gatk_multiple_metrics", mode: "copy"
 
     // NOTE: the method CollectMultipleMetrics has a hidden dependency to R for making plots
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0 r::r=3.6.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0 r::r=3.6.0" : null)
 
     input:
     tuple val(name), val(type), file(bam), file(bai)

modules/04_realignment_around_indels.nf

@@ -13,7 +13,7 @@ process REALIGNMENT_AROUND_INDELS {
     publishDir "${params.output}/${name}/metrics/realignment", mode: "copy", pattern: "*.RA.intervals"
 
     // NOTE: this dependency is fixed to GATK 3 as the realignment around indels is not anymore maintained in GATK 4
-    // but still for some reason for GATK 3 to work the dependency to GATK 4 is needed
+    // but still for some reason for GATK 3 to work the dependency to GATK 4.2.0.0 is needed
     conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0 bioconda::gatk=3.8" : null)
 
     input:

modules/05_bqsr.nf

@@ -11,7 +11,7 @@ process BQSR {
     publishDir "${params.output}/${name}", mode: "copy"
     tag "${name}"
 
-    conda (params.enable_conda ? "bioconda::gatk4=4.2.0.0" : null)
+    conda (params.enable_conda ? "bioconda::gatk4=4.2.5.0" : null)
 
     input:
     tuple val(name), val(type), file(bam), file(bai)

nextflow.config

@@ -46,7 +46,7 @@ process.shell = ['/bin/bash', '-euo', 'pipefail']
 
 cleanup = true
 
-VERSION = '1.7.3'
+VERSION = '1.8.0'
 DOI = 'https://zenodo.org/badge/latestdoi/358400957'
 
 manifest {