From e8027638c051b2d3a47badb9e5dbd1a2d96eff4e Mon Sep 17 00:00:00 2001
From: jsabban <jules.sabban@inrae.fr>
Date: Tue, 12 Sep 2023 11:28:15 +0200
Subject: [PATCH] New step to merge lane fastq files
Ref: #79
---
conf/base.config | 6 ++++++
docs/usage.md | 4 ++++
modules/local/module_core.nf | 26 ++++++++++++++++++++++++++
nextflow.config | 1 +
sub-workflows/local/core_pipeline.nf | 21 +++++++++++++++++++--
5 files changed, 56 insertions(+), 2 deletions(-)
diff --git a/conf/base.config b/conf/base.config
index 22bb404..e918941 100644
--- a/conf/base.config
+++ b/conf/base.config
@@ -55,6 +55,12 @@ process {
module = toolsModuleHash['R']
}
+ withName: MERGE_LANES {
+ cpus = 1
+ memory = { 1.GB * task.attempt }
+ time = { 2.h * task.attempt }
+ }
+
withName: ILLUMINA_FILTER {
publishDir = [
path: "${params.outdir}/IlluminaFilter",
diff --git a/docs/usage.md b/docs/usage.md
index 9908d51..10c2427 100644
--- a/docs/usage.md
+++ b/docs/usage.md
@@ -120,6 +120,10 @@ _Default_ : null
The nG6 like description of the analysis.
_Default_ : null
+- **`--merge_lanes`** [bool]
+Merge fastq over the two lanes in CORE pipeline.
+_Default_ : false
+
### Skipping parameters
There are some availlable flags can be set to not run some parts of the pipeline.
- **`--no_subset`** [bool]
diff --git a/modules/local/module_core.nf b/modules/local/module_core.nf
index cf7e9f6..35bd868 100644
--- a/modules/local/module_core.nf
+++ b/modules/local/module_core.nf
@@ -117,6 +117,32 @@ process DUPLICATED_READS {
'''
}
+process MERGE_LANES {
+ tag "$sample"
+
+ input:
+ tuple val(sample), path(reads)
+
+ output:
+ tuple val(sample), path("*.fastq.gz"), emit: fastq
+
+ script:
+ def args = task.ext.args ?: ''
+ """
+ #!/bin/bash
+
+ R1_files=\$(ls *R1*)
+ R2_files=\$(ls *R2*)
+
+ for file in \$R1_files; do
+ zcat \$file >> ${sample}_R1_001.fastq
+ done
+
+ for file in \$R2_files; do
+ zcat \$file >> ${sample}_R2_001.fastq
+ done
+ """
+}
/* --------------------------------------------------------------------
* OLD PROCESS
diff --git a/nextflow.config b/nextflow.config
index d621dab..242b327 100644
--- a/nextflow.config
+++ b/nextflow.config
@@ -12,6 +12,7 @@ params {
fc_id = ""
fc_type = ""
lane = ""
+ merge_lanes = false
data_nature = ""
species = ""
diff --git a/sub-workflows/local/core_pipeline.nf b/sub-workflows/local/core_pipeline.nf
index 272b1f2..d837b87 100644
--- a/sub-workflows/local/core_pipeline.nf
+++ b/sub-workflows/local/core_pipeline.nf
@@ -15,8 +15,9 @@ include {
FASTQC;
FASTQSCREEN;
DUPLICATED_READS;
+ MERGE_LANES;
} from "$baseDir/modules/local/module_core.nf"
-include { GUNZIP } from "${params.shared_modules}/gzip.nf"
+include { GUNZIP; GZIP } from "${params.shared_modules}/gzip.nf"
include { SEQTK_SAMPLE } from "${params.shared_modules}/seqtk.nf"
include { md5sum as MD5SUM } from "${params.shared_modules}/md5sum.nf"
//-------------------------------------------------
@@ -28,9 +29,25 @@ isResume=workflow.resume
//-------------------------------------------------
workflow CORE {
take:
- ch_read
+ ch_fastq
main:
+ // ----------- Lane merging fastq
+ if (params.merge_lanes) {
+ MERGE_LANES(ch_fastq
+ .collect{it[1]}
+ .flatten()
+ .map { $it -> [ ($it.simpleName =~ /(.*)_S\d+_.*/)[0][1] , $it ] }
+ .groupTuple()
+ )
+
+ GZIP(MERGE_LANES.out.fastq)
+
+ ch_read = GZIP.out
+ } else {
+ ch_read = ch_fastq
+ }
+
// ----------- md5sum
MD5SUM(ch_read.collect{it[1]}.flatten().collect(), params.run_name+'_fastq')
--
GitLab