Minor changes to parameter docs in schema and Assigning ch_input_for_…

…mapping without overwriting reads channel.
nf-core · Sep 29, 2023 · 4810578 · 4810578
1 parent efd07f2
commit 4810578
Show file tree

Hide file tree

Showing 2 changed files with 26 additions and 14 deletions.
diff --git a/nextflow_schema.json b/nextflow_schema.json
@@ -478,16 +478,16 @@
             "properties": {
                 "run_fastq_sharding": {
                     "type": "boolean",
-                    "description": "Turn on sharding to split reads into smaller chunks before mapping",
+                    "description": "Turn on FastQ sharding.",
                     "fa_icon": "fas fa-power-off",
-                    "help_text": "Sharding reads before mapping can help parallelise mapping for very large FASTQs.  "
+                    "help_text": "Sharding will split the FastQs into smaller chunks before mapping. These chunks are then mapped in parallel. This approach can speed up the mapping process for larger FastQ files."
                 },
                 "fastq_shard_size": {
                     "type": "integer",
                     "default": 1000000,
                     "description": "Specify the number of reads in each shard when splitting.",
                     "fa_icon": "fas fa-arrows-alt-v",
-                    "help_text": "Make sure to choose a value that makes sense for your dataset. Small values can create many files. Needs `--run_fastq_sharding`"
+                    "help_text": "Make sure to choose a value that makes sense for your dataset. Small values can create many files, which can end up negatively affecting the overall speed of the mapping process."
                 },
                 "mapping_tool": {
                     "type": "string",

diff --git a/subworkflows/local/map.nf b/subworkflows/local/map.nf
@@ -27,7 +27,7 @@ workflow MAP {
         SEQKIT_SPLIT2( ch_input_for_sharding )
         ch_versions        = ch_versions.mix ( SEQKIT_SPLIT2.out.versions.first() )
 
-        reads = SEQKIT_SPLIT2.out.reads
+        sharded_reads = SEQKIT_SPLIT2.out.reads
             .transpose()
             .map {
                 meta, reads ->
@@ -36,18 +36,30 @@ workflow MAP {
                     [ new_meta, reads ] 
             }
             .groupTuple()
+
+        ch_input_for_mapping = sharded_reads
+            .combine(index)
+            .multiMap {
+                meta, reads, meta2, index ->
+                    new_meta = meta.clone()
+                    new_meta.reference = meta2.id  
+                    reads: [ new_meta, reads ] 
+                    index: [ meta2, index ]
+            }
 
-    }
+    } else {
+
+        ch_input_for_mapping = reads
+            .combine(index)
+            .multiMap {
+                meta, reads, meta2, index ->
+                    new_meta = meta.clone()
+                    new_meta.reference = meta2.id  
+                    reads: [ new_meta, reads ] 
+                    index: [ meta2, index ]
+            }
 
-    ch_input_for_mapping = reads
-        .combine(index)
-        .multiMap {
-            meta, reads, meta2, index ->
-                new_meta = meta.clone()
-                new_meta.reference = meta2.id  
-                reads: [ new_meta, reads ] 
-                index: [ meta2, index ]
-        }
+    }
 
     if ( params.mapping_tool == 'bwaaln' ) {
         FASTQ_ALIGN_BWAALN ( ch_input_for_mapping.reads, ch_input_for_mapping.index )