Merge pull request #9 from TRON-Bioinformatics/rna-support

RNA support through STAR two pass mode

Author: priesgo

.github/workflows/automated_tests.yml

@@ -19,4 +19,5 @@ jobs:
         wget -qO- https://get.nextflow.io | bash && cp nextflow /usr/local/bin/nextflow
     - name: Run tests
       run: |
+        export NXF_VER=22.04.5
         make

Makefile

@@ -21,3 +21,4 @@ test:
 	# STAR indices take over 1 GB and we did not manage to make it work in GitHub actions
 	#bash tests/run_test_11.sh
 	#bash tests/run_test_12.sh
+	#bash tests/run_test_13.sh

README.md

@@ -12,7 +12,7 @@ somatic variant calling.
 Find the documentation here [![Documentation Status](https://readthedocs.org/projects/tronflow-docs/badge/?version=latest)](https://tronflow-docs.readthedocs.io/en/latest/?badge=latest)
 
 This pipeline aligns paired and single end FASTQ files with BWA aln and mem algorithms and with BWA mem 2.
-For RNA-seq star is also supported.
+For RNA-seq STAR is also supported. To increase sensitivity of novel junctions use `--star_two_pass_mode` (recommended for RNAseq variant calling).
 It also includes an initial step of read trimming using FASTP.
 
 
@@ -55,6 +55,8 @@ Optional input:
     * memory: determines the memory required by each job (default: 32g)
     * inception: if enabled it uses an inception, only valid for BWA aln, it requires a fast file system such as flash (default: false)
     * skip_trimming: skips the read trimming step
+    * star_two_pass_mode: activates STAR two-pass mode, increasing sensitivity of novel junction discovery, recommended for RNA variant calling (default: false)
+    * additional_args: additional alignment arguments, only effective in BWA mem, BWA mem 2 and STAR (default: none) 
 
 Output:
     * A BAM file \${name}.bam and its index

modules/02_bwa_mem.nf

@@ -16,7 +16,7 @@ process BWA_MEM {
       file("software_versions.${task.process}.txt")
 
     """
-    bwa mem -t ${task.cpus} ${params.reference} ${fastq1} ${fastq2} | samtools view -uS - | samtools sort - > ${name}.bam
+    bwa mem ${params.additional_args} -t ${task.cpus} ${params.reference} ${fastq1} ${fastq2} | samtools view -uS - | samtools sort - > ${name}.bam
 
     echo ${params.manifest} >> software_versions.${task.process}.txt
     echo "bwa=0.7.17"  >> software_versions.${task.process}.txt
@@ -42,7 +42,7 @@ process BWA_MEM_SE {
       file("software_versions.${task.process}.txt")
 
     """
-    bwa mem -t ${task.cpus} ${params.reference} ${fastq} | samtools view -uS - | samtools sort - > ${name}.bam
+    bwa mem ${params.additional_args} -t ${task.cpus} ${params.reference} ${fastq} | samtools view -uS - | samtools sort - > ${name}.bam
 
     echo ${params.manifest} >> software_versions.${task.process}.txt
     echo "bwa=0.7.17"  >> software_versions.${task.process}.txt

modules/02_bwa_mem_2.nf

@@ -16,7 +16,7 @@ process BWA_MEM_2 {
       file("software_versions.${task.process}.txt")
 
     """
-    bwa-mem2 mem -t ${task.cpus} ${params.reference} ${fastq1} ${fastq2} | samtools view -uS - | samtools sort - > ${name}.bam
+    bwa-mem2 mem ${params.additional_args} -t ${task.cpus} ${params.reference} ${fastq1} ${fastq2} | samtools view -uS - | samtools sort - > ${name}.bam
 
     echo ${params.manifest} >> software_versions.${task.process}.txt
     bwa-mem2 version  >> software_versions.${task.process}.txt
@@ -42,7 +42,7 @@ process BWA_MEM_2_SE {
       file("software_versions.${task.process}.txt")
 
     """
-    bwa-mem2 mem -t ${task.cpus} ${params.reference} ${fastq} | samtools view -uS - | samtools sort - > ${name}.bam
+    bwa-mem2 mem ${params.additional_args} -t ${task.cpus} ${params.reference} ${fastq} | samtools view -uS - | samtools sort - > ${name}.bam
 
     echo ${params.manifest} >> software_versions.${task.process}.txt
     bwa-mem2 version  >> software_versions.${task.process}.txt

modules/02_star.nf

@@ -15,8 +15,10 @@ process STAR {
       tuple val("${name}"), file("${name}.bam"), emit: bams
       file("software_versions.${task.process}.txt")
 
+    script:
+    two_pass_mode_param = params.star_two_pass_mode ? "--twopassMode Basic" : ""
     """
-    STAR --genomeDir ${params.reference} \
+    STAR --genomeDir ${params.reference} ${two_pass_mode_param} ${params.additional_args} \
     --readFilesCommand "gzip -d -c -f" \
     --readFilesIn ${fastq1} ${fastq2} \
     --outSAMmode Full \
@@ -51,8 +53,10 @@ process STAR_SE {
       tuple val("${name}"), file("${name}.bam"), emit: bams
       file("software_versions.${task.process}.txt")
 
+    script:
+    two_pass_mode_param = params.star_two_pass_mode ? "--twopassMode Basic" : ""
     """
-    STAR --genomeDir ${params.reference} \
+    STAR --genomeDir ${params.reference} ${two_pass_mode_param} ${params.additional_args} \
     --readFilesCommand "gzip -d -c -f" \
     --readFilesIn ${fastq} \
     --outSAMmode Full \

nextflow.config

@@ -16,6 +16,8 @@ params.cpus = 8
 params.memory = "32g"
 params.inception = false
 params.skip_trimming = false
+params.star_two_pass_mode = false
+params.additional_args = ""
 
 profiles {
   conda {
@@ -44,7 +46,7 @@ process.shell = ['/bin/bash', '-euo', 'pipefail']
 
 cleanup = true
 
-VERSION = "1.7.0"
+VERSION = "1.8.0"
 
 manifest {
   name = 'TRON-Bioinformatics/tronflow-alignment'
@@ -58,7 +60,7 @@ manifest {
 params.manifest = manifest
 
 params.help_message = """
-TronFlow BWA v${VERSION}
+TronFlow alignment v${VERSION}
 
 Usage:
     nextflow main.nf --input_files input_files [--reference reference.fasta]
@@ -81,6 +83,8 @@ Optional input:
     * memory: determines the memory required by each job (default: 8g)
     * inception: if enabled it uses an inception, only valid for BWA aln, it requires a fast file system such as flash (default: false)
     * skip_trimming: skips the read trimming step
+    * star_two_pass_mode: activates STAR two-pass mode, increasing sensitivity of novel junction discovery, recommended for RNA variant calling (default: false)
+    * additional_args: additional alignment arguments, only effective in BWA mem, BWA mem 2 and STAR (default: none)
 
 Output:
     * A BAM file \${name}.bam and its index

tests/run_test_10.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-output_folder=output/test3
+output_folder=output/test10
 nextflow main.nf -profile test,conda --library single --algorithm mem2 --output $output_folder
 test -s $output_folder/TESTX_S1_L001.bam || { echo "Missing test 3 output file!"; exit 1; }
 test -s $output_folder/TESTX_S1_L001.bam.bai || { echo "Missing test 3 output file!"; exit 1; }

tests/run_test_11.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-output_folder=output/test4
+output_folder=output/test11
 nextflow main.nf -profile test,conda --algorithm star --reference `pwd`/test_data --output $output_folder
 test -s $output_folder/TESTX_S1_L001.bam || { echo "Missing test 4 output file!"; exit 1; }
 test -s $output_folder/TESTX_S1_L001.bam.bai || { echo "Missing test 4 output file!"; exit 1; }

tests/run_test_12.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-output_folder=output/test4
+output_folder=output/test12
 nextflow main.nf -profile test,conda --algorithm star --library single --reference `pwd`/test_data --output $output_folder
 test -s $output_folder/TESTX_S1_L001.bam || { echo "Missing test 4 output file!"; exit 1; }
 test -s $output_folder/TESTX_S1_L001.bam.bai || { echo "Missing test 4 output file!"; exit 1; }