port parabricks/deepvariant to nf-test

.github/conda_skip.yml

@@ -182,6 +182,8 @@ exclude:
     path: modules/nf-core/xeniumranger/resegment
   - profile: conda
     path: modules/nf-core/xeniumranger/import-segmentation
+  - profile: conda
+    path: modules/nf-core/parabricks/deepvariant
   - profile: conda
     path: modules/nf-core/parabricks/mutectcaller
  - profile: conda

modules/nf-core/parabricks/applybqsr/meta.yml

@@ -88,3 +88,4 @@ authors:
   - "@bsiranosian"
 maintainers:
   - "@bsiranosian"
+  - "@famosab"

modules/nf-core/parabricks/dbsnp/meta.yml

@@ -50,3 +50,5 @@ output:
           pattern: "versions.yml"
 authors:
   - "@Furentsu"
+maintainers:
+  - "@famosab"

modules/nf-core/parabricks/deepvariant/main.nf

@@ -1,40 +1,41 @@
 process PARABRICKS_DEEPVARIANT {
     tag "$meta.id"
     label 'process_high'
+    label 'process_gpu'
+    stageInMode 'copy' // needed by the module to work properly can be removed when fixed upstream - Issue #7226
 
-    container "nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1"
+    container "nvcr.io/nvidia/clara/clara-parabricks:4.4.0-1"
 
     input:
     tuple val(meta), path(input), path(input_index), path(interval_file)
     tuple val(ref_meta), path(fasta)
 
     output:
-    tuple val(meta), path("*.vcf"), emit: vcf
-    path "versions.yml",            emit: versions
+    tuple val(meta), path("*.vcf"),      optional: true, emit: vcf
+    tuple val(meta), path("*.g.vcf.gz"), optional: true, emit: gvcf
+    path "versions.yml",                                 emit: versions
 
     when:
     task.ext.when == null || task.ext.when
 
     script:
-
     // Exit if running this module with -profile conda / -profile mamba
     if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
         exit 1, "Parabricks module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
-
     def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
-    def output_file = args =~ "gvcf" ? "${prefix}.g.vcf" : "${prefix}.vcf"
+    def output_file = ("--gvcf" =~ task.ext.args)? "${prefix}.g.vcf.gz" : "${prefix}.vcf"
     def interval_file_command = interval_file ? interval_file.collect{"--interval-file $it"}.join(' ') : ""
+    def num_gpus = task.accelerator ? "--num-gpus $task.accelerator.request" : ''
     """
-
     pbrun \\
         deepvariant \\
         --ref $fasta \\
         --in-bam $input \\
         --out-variants $output_file \\
         $interval_file_command \\
-        --num-gpus $task.accelerator.request \\
+        $num_gpus \\
         $args
 
     cat <<-END_VERSIONS > versions.yml
@@ -44,11 +45,10 @@ process PARABRICKS_DEEPVARIANT {
     """
 
     stub:
-    def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
-    def output_file = args =~ "gvcf" ? "${prefix}.g.vcf" : "${prefix}.vcf"
+    def output_cmd = ("--gvcf" =~ task.ext.args)? "echo '' | gzip > ${prefix}.g.vcf.gz" : "touch ${prefix}.vcf"
     """
-    touch $output_file
+    $output_cmd
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/nf-core/parabricks/deepvariant/meta.yml

@@ -53,12 +53,24 @@ output:
             e.g. [ id:'test' ]
       - "*.vcf":
           type: file
-          description: Variant file.
+          description: vcf file created with deepvariant (does not support .gz for normal vcf), optional
           pattern: "*.vcf"
+  - gvcf:
+      - meta:          
+          type: map
+          description: |
+            Groovy Map containing sample information.
+            e.g. [ id:'test' ]
+      - "*.g.vcf.gz":
+          type: file
+          description: bgzipped gvcf created with deepvariant, optional
+          pattern: "*.g.vcf.gz"
   - versions:
       - versions.yml:
           type: file
           description: File containing software versions.
           pattern: "versions.yml"
 authors:
   - "@bsiranosian"
+maintainers:
+  - "@famosab"

modules/nf-core/parabricks/deepvariant/tests/main.nf.test

@@ -0,0 +1,223 @@
+nextflow_process {
+
+    name "Test Process PARABRICKS_DEEPVARIANT"
+    script "../main.nf"
+    process "PARABRICKS_DEEPVARIANT"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "parabricks"
+    tag "parabricks/deepvariant"
+    tag "gpu"
+
+    test("human - bam") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    [], // bam.bai not needed unless using intervals
+                    []
+                ]
+                input[1] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    path(process.out.vcf[0][1]).vcf.variantsMD5,
+                    process.out.versions,
+                    path(process.out.versions[0]).yaml
+                    ).match() }
+            )
+        }
+
+    }
+
+    test("human - bam - intervals") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
+                ]
+                input[1] = [
+                    [ id:'ref'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    path(process.out.vcf[0][1]).vcf.variantsMD5,
+                    process.out.versions,
+                    path(process.out.versions[0]).yaml
+                    ).match() }
+            )
+        }
+
+    }
+
+    test("human - bam - gvcf") {
+
+        config './nextflow.config'
+
+        when {
+            params {
+                 module_args = '--gvcf'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    [], // bam.bai not needed unless using intervals
+                    []
+                ]
+                input[1] = [
+                    [ id:'ref'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    path(process.out.gvcf[0][1]).vcf.variantsMD5,
+                    process.out.versions,
+                    path(process.out.versions[0]).yaml
+                    ).match()
+                }
+            )
+        }
+
+    }
+
+    test("human - bam - intervals - gvcf") {
+
+        config './nextflow.config'
+
+        when {
+            params {
+                 module_args = '--gvcf'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
+                ]
+                input[1] = [
+                    [ id:'ref'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    path(process.out.gvcf[0][1]).vcf.variantsMD5,
+                    process.out.versions,
+                    path(process.out.versions[0]).yaml
+                    ).match()
+                }
+            )
+        }
+
+    }
+
+    test("human - bam - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    [], // bam.bai not needed unless using intervals
+                    []
+                ]
+                input[1] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    process.out,
+                    path(process.out.versions[0]).yaml
+                    ).match() }
+            )
+        }
+
+    }
+
+    test("human - bam - intervals - gvcf - stub") {
+
+        config './nextflow.config'
+        options "-stub"
+
+        when {
+            params {
+                 module_args = '--gvcf'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
+                ]
+                input[1] = [
+                    [ id:'test'],
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    process.out,
+                    path(process.out.versions[0]).yaml
+                    ).match() }
+            )
+        }
+
+    }
+
+}