From 8de95b793905e0b7a7b798f6b84bcfd316335896 Mon Sep 17 00:00:00 2001
From: Jonathan Manning <jonathan.manning@seqera.io>
Date: Fri, 22 Dec 2023 14:10:33 +0000
Subject: [PATCH] Provide interface to pseudoaligner kmer length

---
 conf/modules.config  | 7 +++++--
 nextflow.config      | 1 +
 nextflow_schema.json | 9 ++++++++-
 3 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/conf/modules.config b/conf/modules.config
index efef36974..d1576e1e3 100644
--- a/conf/modules.config
+++ b/conf/modules.config
@@ -75,7 +75,10 @@ process {
     }
 
     withName: 'SALMON_INDEX' {
-        ext.args   = params.gencode ? '--gencode' : ''
+        ext.args   = { [
+                params.gencode ? '--gencode' : '',
+                params.pseudo_kmer_size ? "-k ${params.pseudo_kmer_size}": ''
+        ].join(' ').trim() }
         publishDir = [
             path: { "${params.outdir}/genome/index" },
             mode: params.publish_dir_mode,
@@ -84,7 +87,7 @@ process {
     }
 
     withName: 'KALLISTO_INDEX' {
-        ext.args   = { params.gencode ? '--gencode' : '' }
+        ext.args   = params.kmer_size ? "-k ${params.pseudo_kmer_size}" : ''
         publishDir = [
             path: { "${params.outdir}/genome/index" },
             mode: params.publish_dir_mode,
diff --git a/nextflow.config b/nextflow.config
index d15d4435c..366a82a74 100644
--- a/nextflow.config
+++ b/nextflow.config
@@ -59,6 +59,7 @@ params {
     // Alignment
     aligner                    = 'star_salmon'
     pseudo_aligner             = null
+    pseudo_aligner_kmer_size   = 31
     seq_center                 = null
     bam_csi_index              = false
     star_ignore_sjdbgtf        = false
diff --git a/nextflow_schema.json b/nextflow_schema.json
index 60e6585cf..f696c76c9 100644
--- a/nextflow_schema.json
+++ b/nextflow_schema.json
@@ -361,6 +361,13 @@
                     "fa_icon": "fas fa-hamburger",
                     "enum": ["salmon", "kallisto"]
                 },
+                "pseudo_aligner_kmer_size": {
+                    "type": "integer",
+                    "default": 31,
+                    "description": "Kmer length passed to indexing step of pseudoaligners",
+                    "help_text": "Failure to set a good kmer size could cause issues with quantification with Kallisto or Salmon. This is mostly an issue for short reads (<50bp), where the default kmer size of 31 is an problem.",
+                    "fa_icon": "fas fa-ruler-horizontal"
+                },
                 "bam_csi_index": {
                     "type": "boolean",
                     "description": "Create a CSI index for BAM files instead of the traditional BAI index. This will be required for genomes with larger chromosome sizes.",
@@ -397,7 +404,7 @@
                 },
                 "min_mapped_reads": {
                     "type": "number",
-                    "default": 5.0,
+                    "default": 5,
                     "fa_icon": "fas fa-percentage",
                     "description": "Minimum percentage of uniquely mapped reads below which samples are removed from further processing.",
                     "help_text": "Some downstream steps in the pipeline will fail if this threshold is too low."