Fix #827

Author: drpatelh

CHANGELOG.md

@@ -13,14 +13,16 @@ Fixed quite a well hidden bug in the UMI processing mode of the pipeline when us
 
 - [[#824](https://github.com/nf-core/rnaseq/issues/824)] - Add explicit docs for usage of featureCounts in the pipeline
 - [[#825](https://github.com/nf-core/rnaseq/issues/825)] - Pipeline fails due to trimming related removal of all reads from a sample
+- [[#827](https://github.com/nf-core/rnaseq/issues/827)] - Control generation of --output-stats when running umi-tools dedup
 - [[#828](https://github.com/nf-core/rnaseq/issues/828)] - Filter BAM output of UMI-tools dedup before passing to Salmon quant
 - Updated pipeline template to [nf-core/tools 2.4.1](https://github.com/nf-core/tools/releases/tag/2.4.1)
 
 ### Parameters
 
-| Old parameter | New parameter         |
-| ------------- | --------------------- |
-|               | `--min_trimmed_reads` |
+| Old parameter | New parameter            |
+| ------------- | ------------------------ |
+|               | `--min_trimmed_reads`    |
+|               | `--umitools_dedup_stats` |
 
 ## [[3.7](https://github.com/nf-core/rnaseq/releases/tag/3.7)] - 2022-05-03
 

modules.json

@@ -112,7 +112,7 @@
                 "git_sha": "e20e57f90b6787ac9a010a980cf6ea98bd990046"
             },
             "umitools/dedup": {
-                "git_sha": "f425aa3cea10015fe9b345b9d6dcc2336b53155f"
+                "git_sha": "84f02a08a5d6d881cb18c31fb15faca4e735a1f2"
             },
             "umitools/extract": {
                 "git_sha": "e745e167c1020928ef20ea1397b6b4d230681b4d"

modules/nf-core/modules/umitools/dedup/main.nf

@@ -28,6 +28,7 @@ params {
     with_umi                   = false
     skip_umi_extract           = false
     umitools_extract_method    = 'string'
+    umitools_dedup_stats       = false
     umitools_bc_pattern        = null
     umi_discard_read           = null
     save_umi_intermeds         = false

modules/nf-core/modules/umitools/dedup/meta.yml

@@ -78,6 +78,12 @@
                     "help_text": "More details can be found in the [UMI-tools documentation](https://umi-tools.readthedocs.io/en/latest/reference/extract.html#extract-method).",
                     "description": "The UMI barcode pattern to use e.g. 'NNNNNN' indicates that the first 6 nucleotides of the read are from the UMI."
                 },
+                "umitools_dedup_stats": {
+                    "type": "boolean",
+                    "fa_icon": "fas fa-barcode",
+                    "help_text": "It can be quite time consuming generating these output stats - see [#827](https://github.com/nf-core/rnaseq/issues/827).",
+                    "description": "Generate output stats when running \"umi_tools dedup\"."
+                },
                 "umi_discard_read": {
                     "type": "integer",
                     "fa_icon": "fas fa-barcode",

nextflow.config

@@ -346,7 +346,8 @@ workflow RNASEQ {
         if (params.with_umi) {
             // Deduplicate genome BAM file before downstream analysis
             DEDUP_UMI_UMITOOLS_GENOME (
-                ch_genome_bam.join(ch_genome_bam_index, by: [0])
+                ch_genome_bam.join(ch_genome_bam_index, by: [0]),
+                params.umitools_dedup_stats
             )
             ch_genome_bam        = DEDUP_UMI_UMITOOLS_GENOME.out.bam
             ch_genome_bam_index  = DEDUP_UMI_UMITOOLS_GENOME.out.bai
@@ -367,7 +368,8 @@ workflow RNASEQ {
 
             // Deduplicate transcriptome BAM file before read counting with Salmon
             DEDUP_UMI_UMITOOLS_TRANSCRIPTOME (
-                ch_transcriptome_sorted_bam.join(ch_transcriptome_sorted_bai, by: [0])
+                ch_transcriptome_sorted_bam.join(ch_transcriptome_sorted_bai, by: [0]),
+                params.umitools_dedup_stats
             )
 
             // Name sort BAM before passing to Salmon