Add par_regions_bed input to DeepVariant (nf-core#6022)

* Add par_regions_bed input to DeepVariant * Add to meta.yml
atrigila · Aug 13, 2024 · a7e8b8a · a7e8b8a
1 parent e21abde
commit a7e8b8a
Show file tree

Hide file tree

Showing 5 changed files with 158 additions and 1 deletion.
diff --git a/modules/nf-core/deepvariant/main.nf b/modules/nf-core/deepvariant/main.nf
@@ -12,6 +12,7 @@ process DEEPVARIANT {
     tuple val(meta2), path(fasta)
     tuple val(meta3), path(fai)
     tuple val(meta4), path(gzi)
+    tuple val(meta5), path(par_bed)
 
     output:
     tuple val(meta), path("${prefix}.vcf.gz")      ,  emit: vcf
@@ -31,6 +32,7 @@ process DEEPVARIANT {
     def args = task.ext.args ?: ''
     prefix = task.ext.prefix ?: "${meta.id}"
     def regions = intervals ? "--regions=${intervals}" : ""
+    def par_regions = par_bed ? "--par_regions_bed=${par_bed}" : ""
     // WARN https://github.com/nf-core/modules/pull/5801#issuecomment-2194293755
     // FIXME Revert this on next version bump
     def VERSION = '1.6.1'
@@ -43,6 +45,7 @@ process DEEPVARIANT {
         --output_gvcf=${prefix}.g.vcf.gz \\
         ${args} \\
         ${regions} \\
+        ${par_regions} \\
         --intermediate_results_dir=tmp \\
         --num_shards=${task.cpus}
 

diff --git a/modules/nf-core/deepvariant/meta.yml b/modules/nf-core/deepvariant/meta.yml
@@ -57,6 +57,15 @@ input:
       type: file
       description: GZI index of reference fasta file
       pattern: "*.gzi"
+  - meta5:
+      type: map
+      description: |
+        Groovy Map containing reference information
+        e.g. [ id:'genome' ]
+  - par_bed:
+      type: file
+      description: BED file containing PAR regions
+      pattern: "*.bed"
 output:
   - meta:
       type: map

diff --git a/modules/nf-core/deepvariant/tests/main.nf.test b/modules/nf-core/deepvariant/tests/main.nf.test
@@ -31,6 +31,9 @@ nextflow_process {
                 input[3] = [
                     [],[]
                 ]
+                input[4] = [
+                    [],[]
+                ]
                 """
             }
         }
@@ -66,6 +69,48 @@ nextflow_process {
                 input[3] = [
                     [],[]
                 ]
+                input[4] = [
+                    [],[]
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+
+    test("homo_sapiens - [cram, crai, genome_bed] - fasta - fai - par_bed") {
+        config "./nextflow-non-autosomal-calling.config"
+        tag "test"
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/illumina/cram/test.paired_end.sorted.cram', checkIfExists: true),
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/illumina/cram/test.paired_end.sorted.cram.crai', checkIfExists: true),
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/genome/genome.bed', checkIfExists: true)
+                ]
+                input[1] = [
+                    [ id:'genome'],
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/genome/genome.fasta', checkIfExists: true)
+                ]
+                input[2] = [
+                    [ id:'genome'],
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/genome/genome.fasta.fai', checkIfExists: true)
+                ]
+                input[3] = [
+                    [],[]
+                ]
+                input[4] = [
+                    [ id:'genome'],
+                      file(params.modules_testdata_base_path + '/genomics/homo_sapiens/genome/genome.blacklist_intervals.bed', checkIfExists: true)
+                ]
                 """
             }
         }
@@ -102,6 +147,9 @@ nextflow_process {
                     [ id:'genome'],
                       file(params.modules_testdata_base_path + '/genomics/homo_sapiens/genome/genome.fasta.gz.gzi', checkIfExists: true)
                 ]
+                input[4] = [
+                    [],[]
+                ]
                 """
             }
         }

diff --git a/modules/nf-core/deepvariant/tests/main.nf.test.snap b/modules/nf-core/deepvariant/tests/main.nf.test.snap
@@ -265,5 +265,94 @@
             "nextflow": "24.04.2"
         },
         "timestamp": "2024-07-01T12:09:13.952808655"
+    },
+    "homo_sapiens - [cram, crai, genome_bed] - fasta - fai - par_bed": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.vcf.gz:md5,d2e26d65dbbcea9b087ed191b5c9841c"
+                    ]
+                ],
+                "1": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.vcf.gz.tbi:md5,0801296d0356415bbf1ef8deb4ec84c3"
+                    ]
+                ],
+                "2": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.g.vcf.gz:md5,4fcaa9a8b55730d191382160c2b5bb0a"
+                    ]
+                ],
+                "3": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.g.vcf.gz.tbi:md5,f468e846904733b3231ecf00ef7cd4a2"
+                    ]
+                ],
+                "4": [
+                    "versions.yml:md5,efbdcb1ad96e0209b31bcafedd0017a8"
+                ],
+                "gvcf": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.g.vcf.gz:md5,4fcaa9a8b55730d191382160c2b5bb0a"
+                    ]
+                ],
+                "gvcf_tbi": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.g.vcf.gz.tbi:md5,f468e846904733b3231ecf00ef7cd4a2"
+                    ]
+                ],
+                "vcf": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.vcf.gz:md5,d2e26d65dbbcea9b087ed191b5c9841c"
+                    ]
+                ],
+                "vcf_tbi": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test_out.vcf.gz.tbi:md5,0801296d0356415bbf1ef8deb4ec84c3"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,efbdcb1ad96e0209b31bcafedd0017a8"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.0",
+            "nextflow": "24.04.3"
+        },
+        "timestamp": "2024-07-23T14:29:24.939680679"
     }
-}
+}
diff --git a/modules/nf-core/deepvariant/tests/nextflow-non-autosomal-calling.config b/modules/nf-core/deepvariant/tests/nextflow-non-autosomal-calling.config
@@ -0,0 +1,8 @@
+process {
+
+    withName: DEEPVARIANT {
+         ext.args = '--model_type=WGS --haploid_contigs chr22'
+         ext.prefix = { "${meta.id}_out" }
+     }
+
+}