From 2b631730a27f1395d192aada2f2d36863e17011d Mon Sep 17 00:00:00 2001
From: JoseEspinosa <kadomu@gmail.com>
Date: Tue, 23 Jul 2024 16:10:33 +0200
Subject: [PATCH] Pass read group fields to bowtie2

---
 CHANGELOG.md                      | 1 +
 conf/modules.config               | 7 ++++++-
 subworkflows/local/input_check.nf | 4 ++--
 3 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index adf84b3d..c33bd644 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -28,6 +28,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [[PR #392](https://github.com/nf-core/chipseq/pull/392)] - Adding line numbers to warnings/errors messages in `bin/check_samplesheet.py`.
 - [[#396](https://github.com/nf-core/chipseq/issues/396)] - Check that samplesheet samples IDs do only have alphanumeric characters, dots, dashes or underscores.
 - [[#378](https://github.com/nf-core/chipseq/issues/378)] - Switch from macs2 to macs3.
+- [[#347](https://github.com/nf-core/chipseq/issues/347)] - Add read group tag to bam files processed by bowtie2.
 
 ### Software dependencies
 
diff --git a/conf/modules.config b/conf/modules.config
index 2251e0ff..22ae4ec0 100644
--- a/conf/modules.config
+++ b/conf/modules.config
@@ -225,7 +225,12 @@ if (params.aligner == 'bwa') {
 if (params.aligner == 'bowtie2') {
     process {
         withName: 'BOWTIE2_ALIGN' {
-            ext.args   = ''
+            ext.args   = {
+                [
+                    meta.read_group ? "--rg-id ${meta.id} --rg SM:${meta.id - ~/_T\d+$/} --rg PL:ILLUMINA --rg LB:${meta.id} --rg PU:1" : '',
+                    params.seq_center ? "--rg CN:${params.seq_center}" : ''
+                ].join(' ').trim()
+            }
             ext.prefix = { "${meta.id}.Lb" }
             publishDir = [
                 [
diff --git a/subworkflows/local/input_check.nf b/subworkflows/local/input_check.nf
index 648a2971..40e20bd1 100644
--- a/subworkflows/local/input_check.nf
+++ b/subworkflows/local/input_check.nf
@@ -29,9 +29,9 @@ def create_fastq_channel(LinkedHashMap row, String seq_center) {
     meta.antibody   = row.antibody
     meta.control    = row.control
 
-    def read_group = "\'@RG\\tID:${meta.id}\\tSM:${meta.id.split('_')[0..-2].join('_')}\\tPL:ILLUMINA\\tLB:${meta.id}\\tPU:1\'"
+    def read_group = "\'@RG\\tID:${meta.id}\\tSM:${meta.id - ~/_T\d+$/}\\tPL:ILLUMINA\\tLB:${meta.id}\\tPU:1\'"
     if (seq_center) {
-        read_group = "\'@RG\\tID:${meta.id}\\tSM:${meta.id.split('_')[0..-2].join('_')}\\tPL:ILLUMINA\\tLB:${meta.id}\\tPU:1\\tCN:${seq_center}\'"
+        read_group = "\'@RG\\tID:${meta.id}\\tSM:${meta.id - ~/_T\d+$/}\\tPL:ILLUMINA\\tLB:${meta.id}\\tPU:1\\tCN:${seq_center}\'"
     }
     meta.read_group = read_group