Added an output channel for the gzi index (#2036)

* When the `-i` option of bgzip is given, an index is created and needs to be considered an output

* It is not necessary to use a subfolder

* Updated meta.yml too

* Also check that an index name hasn't been requested

Co-authored-by: FriederikeHanssen <Friederike.hanssen@qbic.uni-tuebingen.de>

modules/tabix/bgzip/main.nf

@@ -12,6 +12,7 @@ process TABIX_BGZIP {
 
     output:
     tuple val(meta), path("${prefix}*"), emit: output
+    tuple val(meta), path("*gzi")      , emit: gzi, optional: true
     path  "versions.yml"               , emit: versions
 
     when:
@@ -23,6 +24,10 @@ process TABIX_BGZIP {
     in_bgzip = input.toString().endsWith(".gz")
     command1 = in_bgzip ? '-d' : '-c'
     command2 = in_bgzip ? ''   : " > ${prefix}.${input.getExtension()}.gz"
+    // Name the index according to $prefix, unless a name has been requested
+    if ((args.matches("(^| )-i\\b") || args.matches("(^| )--index(\$| )")) && !args.matches("(^| )-I\\b") && !args.matches("(^| )--index-name\\b")) {
+        args = args + " -I ${prefix}.${input.getExtension()}.gz.gzi"
+    }
     """
     bgzip $command1 $args -@${task.cpus} $input $command2
 

modules/tabix/bgzip/meta.yml

@@ -32,6 +32,10 @@ output:
       type: file
       description: Output compressed/decompressed file
       pattern: "*."
+  - gzi:
+      type: file
+      description: Optional gzip index file for compressed inputs
+      pattern: "*.gzi"
   - versions:
       type: file
       description: File containing software versions

tests/modules/tabix/bgzip/main.nf

@@ -3,6 +3,7 @@
 nextflow.enable.dsl = 2
 
 include { TABIX_BGZIP } from '../../../../modules/tabix/bgzip/main.nf'
+include { TABIX_BGZIP as TABIX_BGZIP_WITH_GZI } from '../../../../modules/tabix/bgzip/main.nf'
 
 workflow test_tabix_bgzip_compress {
     input = [ [ id:'test' ], // meta map
@@ -12,6 +13,14 @@ workflow test_tabix_bgzip_compress {
     TABIX_BGZIP ( input )
 }
 
+workflow test_tabix_bgzip_compress_gzi {
+    input = [ [ id:'test' ], // meta map
+              [ file(params.test_data['sarscov2']['illumina']['test_vcf'], checkIfExists: true) ]
+            ]
+
+    TABIX_BGZIP_WITH_GZI ( input )
+}
+
 workflow test_tabix_bgzip_decompress {
     input = [ [ id:'test' ], // meta map
               [ file(params.test_data['sarscov2']['genome']['test_bed_gz'], checkIfExists: true) ]

tests/modules/tabix/bgzip/nextflow.config

@@ -2,4 +2,7 @@ process {
 
     publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
 
+    withName: TABIX_BGZIP_WITH_GZI {
+        ext.args = ' -i'
+    }
 }

tests/modules/tabix/bgzip/test.yml

@@ -6,6 +6,18 @@
   files:
     - path: ./output/tabix/test.vcf.gz
       md5sum: fc178eb342a91dc0d1d568601ad8f8e2
+    - path: ./output/tabix/test.vcf.gz.gzi
+      should_exist: false
+- name: tabix bgzip compress gzi
+  command: nextflow run ./tests/modules/tabix/bgzip -entry test_tabix_bgzip_compress_gzi -c ./tests/config/nextflow.config -c ./tests/modules/tabix/bgzip/nextflow.config
+  tags:
+    - tabix
+    - tabix/bgzip
+  files:
+    - path: ./output/tabix/test.vcf.gz
+      md5sum: fc178eb342a91dc0d1d568601ad8f8e2
+    - path: ./output/tabix/test.vcf.gz.gzi
+      md5sum: 7dea362b3fac8e00956a4952a3d4f474
 - name: tabix bgzip decompress
   command: nextflow run ./tests/modules/tabix/bgzip -entry test_tabix_bgzip_decompress -c ./tests/config/nextflow.config -c ./tests/modules/tabix/bgzip/nextflow.config
   tags: