References can be called manually from samplesheet

DOH-JDJ0303 · Jul 10, 2024 · 260ccdf · 260ccdf
1 parent 2461b6a
commit 260ccdf
Show file tree

Hide file tree

Showing 3 changed files with 38 additions and 7 deletions.
diff --git a/subworkflows/local/classify.nf b/subworkflows/local/classify.nf
@@ -16,12 +16,15 @@ include { NCBI_DATASETS                         } from '../../modules/local/ncbi
 
 workflow CLASSIFY {
     take:
-    ch_reads // channel: [ val(meta), path(reads) ]
+    ch_reads // channel: [ val(meta), path(reads), path/val(reference ) ]
     ch_refs  // channel: [ val(meta), path(reference) ]
 
     main:
 
     ch_versions = Channel.empty()
+
+    ch_man_refs = ch_reads.map{ meta, reads, ref -> [ meta, ref] }
+    ch_reads    = ch_reads.map{ meta, reads, ref -> [ meta, reads ] }
 
     /* 
     =============================================================================================================================
@@ -188,7 +191,29 @@ workflow CLASSIFY {
             .set{ ch_ref_list }
 
     }
-
+
+    /* 
+    =============================================================================================================================
+        REFERENCE SELECTION: Manual
+        - Adds list of references provided in the samplesheet
+    =============================================================================================================================
+    */
+    ch_man_refs
+        .filter{ meta, ref -> ref }
+        .map{ meta, ref -> [ meta, ref.tokenize(';') ] }
+        .transpose()
+        .map{ meta, ref -> [ meta, file(ref).exists() ? file(ref).getSimpleName() : ref, file(ref).exists() ? file(ref) : null ] }
+        .set{ ch_man_refs }
+    ch_man_refs
+        .filter{ meta, ref_id, ref -> ! ref }
+        .map{ meta, ref_id, ref -> [ ref_id, meta ] }
+        .combine(ch_refs.map{ meta, segment, ref -> [ file(ref).getSimpleName(), ref ] }, by: 0)
+        .map{ ref_id, meta, ref -> [ meta, ref_id, ref ] }
+        .concat(ch_man_refs.filter{ meta, ref_id, ref -> ref })
+        .concat( ch_ref_list )
+        .unique()
+        .set{ ch_ref_list }
+
     emit:
     ref_list   = ch_ref_list   // channel: [ val(sample_meta), val(ref_id), path(ref_path) ]
     sm_summary = ch_sm_summary // channel: [ val(meta), val(result) ]

diff --git a/subworkflows/local/input_check.nf b/subworkflows/local/input_check.nf
@@ -56,6 +56,9 @@ def create_fastq_channel(LinkedHashMap row) {
     meta.id         = row.sample
     meta.single_end = row.single_end.toBoolean()
 
+    // define manual references
+    ref = row.reference ? row.reference : null
+
     // define validation fields
     truth = row.truth ? file(row.truth, checkIfExists: true) : null
     inter_group = row.inter_group ? row.inter_group : null
@@ -67,12 +70,12 @@ def create_fastq_channel(LinkedHashMap row) {
         exit 1, "ERROR: Please check input samplesheet -> Read 1 FastQ file does not exist!\n${row.fastq_1}"
     }
     if (meta.single_end) {
-        fastq_meta = [ meta, [ file(row.fastq_1) ], truth, inter_group, intra_group ]
+        fastq_meta = [ meta, [ file(row.fastq_1) ], ref, truth, inter_group, intra_group ]
     } else {
         if (!file(row.fastq_2).exists()) {
             exit 1, "ERROR: Please check input samplesheet -> Read 2 FastQ file does not exist!\n${row.fastq_2}"
         }
-        fastq_meta = [ meta, [ file(row.fastq_1), file(row.fastq_2) ], truth, inter_group, intra_group ]
+        fastq_meta = [ meta, [ file(row.fastq_1), file(row.fastq_2) ], ref, truth, inter_group, intra_group ]
     }    
     return fastq_meta
 }

diff --git a/workflows/vaper.nf b/workflows/vaper.nf
@@ -87,7 +87,7 @@ workflow VAPER {
     )
     ch_versions = ch_versions.mix(INPUT_CHECK.out.versions)
 
-    INPUT_CHECK.out.reads.map{ meta, reads, truth, inter_group, intra_group -> [ meta, reads ] }.set{ ch_reads }
+    INPUT_CHECK.out.reads.map{ meta, reads, reference, truth, inter_group, intra_group -> [ meta, reads ] }.set{ ch_reads }
     INPUT_CHECK.out.refs.set{ ch_refs }
 
     /* 
@@ -131,7 +131,10 @@ workflow VAPER {
     // SUBWORKFLOW: Classify viruses
     //
     CLASSIFY (
-        FASTP.out.reads,
+        FASTP
+            .out
+            .reads
+            .join(INPUT_CHECK.out.reads.map{ meta, reads, reference, truth, inter_group, intra_group -> [ meta, reference ] }, by: 0),
         ch_refs
     )
 
@@ -201,7 +204,7 @@ workflow VAPER {
     */
 
     VALIDATE(
-        INPUT_CHECK.out.reads.map{ meta, reads, truth, inter_group, intra_group -> [ meta, truth, inter_group, intra_group ] },
+        INPUT_CHECK.out.reads.map{ meta, reads, ref, truth, inter_group, intra_group -> [ meta, truth, inter_group, intra_group ] },
         ASSEMBLE.out.consensus.map{ meta, ref_id, consensus -> [ meta, consensus ] }
     )