From 42f318fe752f73858ec00e46e555436362b47608 Mon Sep 17 00:00:00 2001
From: drpatelh <drhpatel@gmail.com>
Date: Wed, 9 Dec 2020 21:20:11 +0000
Subject: [PATCH] Update Salmon modules

---
 software/salmon/index/main.nf | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/software/salmon/index/main.nf b/software/salmon/index/main.nf
index 754818fe..ef6ce209 100644
--- a/software/salmon/index/main.nf
+++ b/software/salmon/index/main.nf
@@ -5,7 +5,7 @@ params.options = [:]
 def options    = initOptions(params.options)
 
 process SALMON_INDEX {
-    tag "$fasta"
+    tag "$transcript_fasta"
     label "process_medium"
     publishDir "${params.outdir}",
         mode: params.publish_dir_mode,
@@ -15,21 +15,33 @@ process SALMON_INDEX {
     container "quay.io/biocontainers/salmon:1.3.0--hf69c8f4_0"
 
     input:
-    path fasta
+    path genome_fasta
+    path transcript_fasta
 
     output:
     path "salmon"       , emit: index
     path "*.version.txt", emit: version
 
     script:
-    def software = getSoftwareName(task.process) 
+    def software      = getSoftwareName(task.process)
+    def get_decoy_ids = "grep '^>' $genome_fasta | cut -d ' ' -f 1 > decoys.txt"
+    def gentrome      = "gentrome.fa"
+    if (genome_fasta.endsWith('.gz')) {
+        get_decoy_ids = "grep '^>' <(gunzip -c $genome_fasta) | cut -d ' ' -f 1 > decoys.txt"
+        gentrome      = "gentrome.fa.gz"
+    }
     """
+    $get_decoy_ids
+    sed -i.bak -e 's/>//g' decoys.txt
+    cat $transcript_fasta $genome_fasta > $gentrome
+
     salmon \\
         index \\
         --threads $task.cpus \\
-        -t $fasta \\
+        -t gentrome.fa \\
+        -d decoys.txt \\
         $options.args \\
         -i salmon
     salmon --version | sed -e "s/salmon //g" > ${software}.version.txt
     """
-}
+}
\ No newline at end of file