reduce pipeline to clustering

genepi · Jul 25, 2024 · ad24d5a · ad24d5a
1 parent a1a7097
commit ad24d5a
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 29 deletions.
diff --git a/lib/processes/cluster.nf b/lib/processes/cluster.nf
@@ -1,14 +1,13 @@
 consensus_fasta="consensus.fasta"
 vsearch_dir="vsearch_clusters"
-
+detected_umis_file_name="dected_umis.fastq"
 process CLUSTER {
-
+    publishDir "${params.output}/${sample}/clustering/${type}", pattern: "cluster*", mode: 'copy'
+
     input:
-        tuple val( sample ), val( target ), path( detected_umis_fastq )
-        val ( type )
+        path sample
     output:
-        tuple val( "${sample}" ), val( "${target}" ), path( "${consensus_fasta}" ), optional: true, emit:consensus_fasta
-        tuple val( "${sample}" ), val( "${target}" ), path( "cluster*" ), optional: true, emit:cluster_fastas
+        tuple val( "${sample.baseName}" ), path( "cluster*" ), optional: true, emit:cluster_fastas
 
     script:
         def id = "${type}" == "raw" ? params.vsearch_sequence_identity : 0.99
@@ -20,7 +19,7 @@ process CLUSTER {
         --minseqlength ${params.min_length} \
         --maxseqlength ${params.max_length} \
         --threads ${params.threads} \
-        --cluster_fast ${detected_umis_fastq} \
+        --cluster_fast ${sample}/${detected_umis_file_name} \
         --clusterout_sort \
         --gapopen 0E/5I \
         --gapext 0E/2I \

diff --git a/lib/processes/reformat_filter_cluster.nf b/lib/processes/reformat_filter_cluster.nf
@@ -3,7 +3,7 @@ process REFORMAT_FILTER_CLUSTER {
     publishDir "${params.output}/${sample}/stats/${type}", pattern: "*tsv", mode: 'copy'
 
     input:
-        tuple val( sample ), val( target ), path( cluster )
+        tuple val( sample ), path( cluster )
         val( type )
         path umi_parse_clusters_python
 

diff --git a/lib/workflows/umi-pipeline.nf b/lib/workflows/umi-pipeline.nf
@@ -33,7 +33,7 @@ n_parsed_cluster            = [:]
 // Remove barcode01 and uncalssified from the input fastq folder
 Channel.fromPath("${params.input}/*", type: 'dir')
     .filter( ~/.*barcode(([0-9][0-9]))/ )
-    .set { fastq_files_ch }
+    .set { detected_umis_ch }
 
 ////////////////////
 // BEGIN PIPELINE //
@@ -54,26 +54,7 @@ include {REFORMAT_FILTER_CLUSTER} from '../processes/reformat_filter_cluster.nf'
 workflow UMI_PIPELINE {
 
     main:
-        COPY_BED( bed )
-
-        if( params.subsampling ){
-            MERGE_FASTQ( fastq_files_ch )
-            SUBSAMPLING( MERGE_FASTQ.out.merged_fastq )
-            .set { merged_fastq }
-        } else {
-            MERGE_FASTQ( fastq_files_ch )
-            .set { merged_fastq }
-        }
-
-        merged_fastq
-        .filter { sample, target, fastq_file -> fastq_file.countFastq() > params.min_reads_per_barcode }
-        .set { merged_filtered_fastq }
-
-        MAP_READS( merged_filtered_fastq, raw, reference )
-        SPLIT_READS( MAP_READS.out.bam_consensus, COPY_BED.out.bed, raw, umi_filter_reads )
-        DETECT_UMI_FASTQ( SPLIT_READS.out.split_reads_fastx, raw, umi_extract )
-        CLUSTER( DETECT_UMI_FASTQ.out.umi_extract_fastq, raw )
-
+        CLUSTER( detected_umis_ch)
         REFORMAT_FILTER_CLUSTER( CLUSTER.out.cluster_fastas, raw, umi_parse_clusters )
 }