update to kraken2: breaking change - output channels renamed (#1525)

* updated kraken2 module to include optional classification of each input reads, and make fastq outputs optional
NB: this is a breaking change, because the output channels have been renamed as a consequence of changes

* updated yml

* pigz command made optional, in order to be executed only if fastq of classified/unclassified reads are saved

* updated test yaml file for kraken2

* fixed TODOs and renamed variables and outputs

* untar in conda cannot keep same md5sum of version, and therefore md5sum check removed

* improved description of the options

Co-authored-by: James A. Fellows Yates <jfy133@gmail.com>

modules/kraken2/kraken2/main.nf

@@ -10,11 +10,14 @@ process KRAKEN2_KRAKEN2 {
     input:
     tuple val(meta), path(reads)
     path  db
+    val save_output_fastqs
+    val save_reads_assignment
 
     output:
-    tuple val(meta), path('*classified*')  , emit: classified
+    tuple val(meta), path('*classified*')     , optional:true, emit: classified_reads_fastq
-    tuple val(meta), path('*unclassified*'), emit: unclassified
+    tuple val(meta), path('*unclassified*')   , optional:true, emit: unclassified_reads_fastq
-    tuple val(meta), path('*report.txt')   , emit: txt
+    tuple val(meta), path('*classifiedreads*'), optional:true, emit: classified_reads_assignment
+    tuple val(meta), path('*report.txt')                     , emit: report
     path "versions.yml"                                      , emit: versions
 
     when:
@@ -26,19 +29,25 @@ process KRAKEN2_KRAKEN2 {
     def paired       = meta.single_end ? "" : "--paired"
     def classified   = meta.single_end ? "${prefix}.classified.fastq"   : "${prefix}.classified#.fastq"
     def unclassified = meta.single_end ? "${prefix}.unclassified.fastq" : "${prefix}.unclassified#.fastq"
+    def classified_command = save_output_fastqs ? "--classified-out ${classified}" : ""
+    def unclassified_command = save_output_fastqs ? "--unclassified-out ${unclassified}" : ""
+    def readclassification_command = save_reads_assignment ? "--output ${prefix}.kraken2.classifiedreads.txt" : ""
+    def compress_reads_command = save_output_fastqs ? "pigz -p $task.cpus *.fastq" : ""
+
     """
     kraken2 \\
         --db $db \\
         --threads $task.cpus \\
-        --unclassified-out $unclassified \\
-        --classified-out $classified \\
         --report ${prefix}.kraken2.report.txt \\
         --gzip-compressed \\
+        $unclassified_command \\
+        $classified_command \\
+        $readclassification_command \\
         $paired \\
         $args \\
         $reads
 
-    pigz -p $task.cpus *.fastq
+    $compress_reads_command
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/kraken2/kraken2/meta.yml

@@ -27,25 +27,40 @@ input:
   - db:
       type: directory
       description: Kraken2 database
+  - save_output_fastqs:
+      type: boolean
+      description: |
+        If true, optional commands are added to save classified and unclassified reads
+        as fastq files
+  - save_reads_assignment:
+      type: boolean
+      description: |
+        If true, an optional command is added to save a file reporting the taxonomic
+        classification of each input read
 output:
   - meta:
       type: map
       description: |
         Groovy Map containing sample information
         e.g. [ id:'test', single_end:false ]
-  - classified:
+  - classified_reads_fastq:
       type: file
       description: |
-        Reads classified to belong to any of the taxa
+        Reads classified as belonging to any of the taxa
         on the Kraken2 database.
       pattern: "*{fastq.gz}"
-  - unclassified:
+  - unclassified_reads_fastq:
       type: file
       description: |
-        Reads not classified to belong to any of the taxa
+        Reads not classified to any of the taxa
         on the Kraken2 database.
       pattern: "*{fastq.gz}"
-  - txt:
+  - classified_reads_assignment:
+      type: file
+      description: |
+        Kraken2 output file indicating the taxonomic assignment of
+        each input read
+  - report:
       type: file
       description: |
         Kraken2 report containing stats about classified

tests/modules/kraken2/kraken2/main.nf

@@ -12,7 +12,7 @@ workflow test_kraken2_kraken2_single_end {
     db    = [ [], file(params.test_data['sarscov2']['genome']['kraken2_tar_gz'], checkIfExists: true) ]
 
     UNTAR ( db )
-    KRAKEN2_KRAKEN2 ( input, UNTAR.out.untar.map{ it[1] } )
+    KRAKEN2_KRAKEN2 ( input, UNTAR.out.untar.map{ it[1] }, true, false )
 }
 
 workflow test_kraken2_kraken2_paired_end {
@@ -23,5 +23,15 @@ workflow test_kraken2_kraken2_paired_end {
     db    = [ [], file(params.test_data['sarscov2']['genome']['kraken2_tar_gz'], checkIfExists: true) ]
 
     UNTAR ( db )
-    KRAKEN2_KRAKEN2 ( input, UNTAR.out.untar.map{ it[1] } )
+    KRAKEN2_KRAKEN2 ( input, UNTAR.out.untar.map{ it[1] }, true, false )
+}
+
+workflow test_kraken2_kraken2_classifyreads {
+    input = [ [ id:'test', single_end:true ], // meta map
+              [ file(params.test_data['sarscov2']['illumina']['test_1_fastq_gz'], checkIfExists: true) ]
+            ]
+    db    = [ [], file(params.test_data['sarscov2']['genome']['kraken2_tar_gz'], checkIfExists: true) ]
+
+    UNTAR ( db )
+    KRAKEN2_KRAKEN2 ( input, UNTAR.out.untar.map{ it[1] }, false, true )
 }

tests/modules/kraken2/kraken2/test.yml

@@ -1,29 +1,43 @@
-- name: kraken2 kraken2 single-end
+- name: kraken2 kraken2 test_kraken2_kraken2_single_end
-  command: nextflow run ./tests/modules/kraken2/kraken2 -entry test_kraken2_kraken2_single_end -c ./tests/config/nextflow.config -c ./tests/modules/kraken2/kraken2/nextflow.config
+  command: nextflow run tests/modules/kraken2/kraken2 -entry test_kraken2_kraken2_single_end -c tests/config/nextflow.config
   tags:
     - kraken2
     - kraken2/kraken2
   files:
     - path: output/kraken2/test.classified.fastq.gz
-      should_exist: true
-    - path: output/kraken2/test.unclassified.fastq.gz
-      should_exist: true
     - path: output/kraken2/test.kraken2.report.txt
       md5sum: 4227755fe40478b8d7dc8634b489761e
+    - path: output/kraken2/test.unclassified.fastq.gz
+    - path: output/kraken2/versions.yml
+      md5sum: 6e3ad947ac8dee841a89216071c181cc
+    - path: output/untar/versions.yml
 
-- name: kraken2 kraken2 paired-end
+- name: kraken2 kraken2 test_kraken2_kraken2_paired_end
-  command: nextflow run ./tests/modules/kraken2/kraken2 -entry test_kraken2_kraken2_paired_end -c ./tests/config/nextflow.config -c ./tests/modules/kraken2/kraken2/nextflow.config
+  command: nextflow run tests/modules/kraken2/kraken2 -entry test_kraken2_kraken2_paired_end -c tests/config/nextflow.config
   tags:
     - kraken2
     - kraken2/kraken2
   files:
     - path: output/kraken2/test.classified_1.fastq.gz
-      should_exist: true
     - path: output/kraken2/test.classified_2.fastq.gz
-      should_exist: true
-    - path: output/kraken2/test.unclassified_1.fastq.gz
-      should_exist: true
-    - path: output/kraken2/test.unclassified_2.fastq.gz
-      should_exist: true
     - path: output/kraken2/test.kraken2.report.txt
       md5sum: 4227755fe40478b8d7dc8634b489761e
+    - path: output/kraken2/test.unclassified_1.fastq.gz
+    - path: output/kraken2/test.unclassified_2.fastq.gz
+    - path: output/kraken2/versions.yml
+      md5sum: 604482fe7a4519f890fae9c8beb1bd6e
+    - path: output/untar/versions.yml
+
+- name: kraken2 kraken2 test_kraken2_kraken2_classifyreads
+  command: nextflow run tests/modules/kraken2/kraken2 -entry test_kraken2_kraken2_classifyreads -c tests/config/nextflow.config
+  tags:
+    - kraken2
+    - kraken2/kraken2
+  files:
+    - path: output/kraken2/test.kraken2.classifiedreads.txt
+      md5sum: e7a90531f0d8d777316515c36fe4cae0
+    - path: output/kraken2/test.kraken2.report.txt
+      md5sum: 4227755fe40478b8d7dc8634b489761e
+    - path: output/kraken2/versions.yml
+      md5sum: 3488c304259e83c5bea573403293fce9
+    - path: output/untar/versions.yml