From b8dc66ec8ed9d3755f52ddf42068928de8233bb2 Mon Sep 17 00:00:00 2001
From: Gisela Gabernet <gisela.gabernet@gmail.com>
Date: Wed, 17 Jul 2024 13:35:23 -0400
Subject: [PATCH] add locus selection to rnaseq workflow

---
 conf/modules.config                                    |  8 ++++++++
 ...rsedb_select.nf => changeo_parsedb_select_locus.nf} | 10 +++-------
 subworkflows/local/rnaseq_input.nf                     |  6 +++++-
 3 files changed, 16 insertions(+), 8 deletions(-)
 rename modules/local/changeo/{changeo_parsedb_select.nf => changeo_parsedb_select_locus.nf} (72%)

diff --git a/conf/modules.config b/conf/modules.config
index 3dc63fa9..dd315b85 100644
--- a/conf/modules.config
+++ b/conf/modules.config
@@ -422,6 +422,14 @@ process {
         ]
     }
 
+    withName: CHANGEO_PARSEDB_SELECT_LOCUS {
+        publishDir = [
+            path: { "${params.outdir}/vdj_annotation/select-locus/${meta.id}" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+
     withName: CHANGEO_PARSEDB_SPLIT {
         publishDir = [
             path: { "${params.outdir}/vdj_annotation/04-select-productive/${meta.id}" },
diff --git a/modules/local/changeo/changeo_parsedb_select.nf b/modules/local/changeo/changeo_parsedb_select_locus.nf
similarity index 72%
rename from modules/local/changeo/changeo_parsedb_select.nf
rename to modules/local/changeo/changeo_parsedb_select_locus.nf
index 2bba4916..32805c26 100644
--- a/modules/local/changeo/changeo_parsedb_select.nf
+++ b/modules/local/changeo/changeo_parsedb_select_locus.nf
@@ -1,4 +1,4 @@
-process CHANGEO_PARSEDB_SELECT {
+process CHANGEO_PARSEDB_SELECT_LOCUS {
     tag "$meta.id"
     label 'process_low'
     label 'immcantation'
@@ -18,25 +18,21 @@ process CHANGEO_PARSEDB_SELECT {
     path "versions.yml" , emit: versions
 
     script:
-    def args = task.ext.args ?: ''
-    def args2 = task.ext.args2 ?: ''
     if (meta.locus.toUpperCase() == 'IG'){
         """
-        ParseDb.py select -d $tab $args --outname ${meta.id} > ${meta.id}_select_command_log.txt
+        ParseDb.py select -d $tab -f locus -u "IG[HKL]" --regex --outname ${meta.id} > ${meta.id}_select_command_log.txt
 
         cat <<-END_VERSIONS > versions.yml
         "${task.process}":
-            igblastn: \$( igblastn -version | grep -o "igblast[0-9\\. ]\\+" | grep -o "[0-9\\. ]\\+" )
             changeo: \$( ParseDb.py --version | awk -F' '  '{print \$2}' )
         END_VERSIONS
         """
     } else if (meta.locus.toUpperCase() == 'TR'){
         """
-        ParseDb.py select -d $tab $args2 --outname ${meta.id} > "${meta.id}_command_log.txt"
+        ParseDb.py select -d $tab -f locus -u "TR[ABDG]" --regex --outname ${meta.id} > "${meta.id}_command_log.txt"
 
         cat <<-END_VERSIONS > versions.yml
         "${task.process}":
-            igblastn: \$( igblastn -version | grep -o "igblast[0-9\\. ]\\+" | grep -o "[0-9\\. ]\\+" )
             changeo: \$( ParseDb.py --version | awk -F' '  '{print \$2}' )
         END_VERSIONS
         """
diff --git a/subworkflows/local/rnaseq_input.nf b/subworkflows/local/rnaseq_input.nf
index a04898dd..12253f5a 100644
--- a/subworkflows/local/rnaseq_input.nf
+++ b/subworkflows/local/rnaseq_input.nf
@@ -1,6 +1,7 @@
 include { PREPARE_TRUST4_REFERENCE                                      } from '../../modules/local/prepare_trust4_reference'
 include { TRUST4                                                        } from '../../modules/nf-core/trust4/main'
 include { FASTQ_INPUT_CHECK                                             } from '../../subworkflows/local/fastq_input_check'
+include { CHANGEO_PARSEDB_SELECT_LOCUS                                  } from '../../modules/local/changeo/changeo_parsedb_select_locus'
 include { CHANGEO_CONVERTDB_FASTA as CHANGEO_CONVERTDB_FASTA_FROM_AIRR  } from '../../modules/local/changeo/changeo_convertdb_fasta'
 include { FASTP                                                         } from '../../modules/nf-core/fastp/main'
 include { RENAME_FASTQ as RENAME_FASTQ_TRUST4                           } from '../../modules/local/rename_fastq'
@@ -106,10 +107,13 @@ workflow RNASEQ_INPUT {
     // create channel with airr file
     ch_trust4_airr_file.bulk.mix ( ch_trust4_airr_file.sc ).set { ch_trust4_airr }
 
+    // select only provided locus
+    CHANGEO_PARSEDB_SELECT_LOCUS(ch_trust4_airr)
+
 
     // convert airr tsv to fasta
     CHANGEO_CONVERTDB_FASTA_FROM_AIRR(
-                ch_trust4_airr
+                CHANGEO_PARSEDB_SELECT_LOCUS.out.tab
             )
 
     ch_fasta = CHANGEO_CONVERTDB_FASTA_FROM_AIRR.out.fasta