Hiphase - new module

modules/nf-core/hiphase/environment.yml

@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::hiphase=1.4.5"

modules/nf-core/hiphase/main.nf

@@ -0,0 +1,58 @@
+process HIPHASE {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/hiphase:1.4.5--h9ee0642_0':
+        'biocontainers/hiphase:1.4.5--h9ee0642_0' }"
+
+    input:
+    tuple val(meta), path(vcf)
+    tuple val(meta1), path(csi)
+    tuple val(meta2), path(bam)
+    tuple val(meta3), path(bai)
+    tuple val(meta4), path(fasta)
+
+    output:
+    tuple val(meta), path("*.vcf"), emit: vcf
+    tuple val(meta), path("*.csv"),  emit: csv
+    path "versions.yml"           , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    """
+    hiphase \
+        --bam $bam \
+        --vcf $vcf \
+        --output-vcf ${prefix}.phased.vcf \
+        --output-bam ${prefix}.phased.bam \
+        --reference $fasta \
+        --stats-file ${prefix}.stats.csv \
+        $args \
+        --threads ${task.cpus}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        hiphase: \$(hiphase --version |& sed '1!d ; s/hiphase //')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.phased.vcf
+    touch ${prefix}.stats.csv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        hiphase: \$(hiphase --version |& sed '1!d ; s/hiphase //')
+    END_VERSIONS
+    """
+}

modules/nf-core/hiphase/meta.yml

@@ -0,0 +1,94 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "hiphase"
+description: write your description here
+keywords:
+  - sort
+  - example
+  - genomics
+tools:
+  - "hiphase":
+      description: "Small and structural variant phasing tool for PacBio HiFi reads"
+      homepage: "https://github.com/PacificBiosciences/HiPhase"
+      documentation: "https://github.com/PacificBiosciences/HiPhase"
+      tool_dev_url: "https://github.com/PacificBiosciences/HiPhase"
+      doi: "10.1234/placeholder.doi"
+      licence: ["Pacific Biosciences Software License Agreement"]
+      identifier: biotools:hiphase
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - vcf:
+        type: file
+        description: VCF File
+        pattern: "*.{vcf}"
+  - - meta1:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - csi:
+        type: file
+        description: CSI File associated with VCF
+        pattern: "*.{csi}"
+  - - meta2:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - bam:
+        type: file
+        description: Sorted BAM/CRAM/SAM file
+        pattern: "*.{bam}"
+  - - meta3:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - bai:
+        type: file
+        description: Index File BAI associated with BAM file
+        pattern: "*.{bai}"
+  - - meta4:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - fasta:
+        type: file
+        description: FASTA File (used for the alignment of the BAM file)
+        pattern: "*.{fasta}"
+output:
+  - vcf:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.vcf":
+          type: file
+          description: Phased VCF file
+          pattern: "*.vcf"
+  - csv:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.csv":
+          type: file
+          description: stats associated with the phasing
+          pattern: "*.csv"
+
+  - versions:
+      - "versions.yml":
+          type: file
+          description: File containing software versions
+          pattern: "versions.yml"
+
+authors:
+  - "@tanyasarkjain"
+maintainers:
+  - "@tanyasarkjain"

modules/nf-core/hiphase/tests/main.nf.test

@@ -0,0 +1,98 @@
+nextflow_process {
+    name "Test Process HIPHASE"
+    script "../main.nf"
+    process "HIPHASE"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "hiphase"
+
+    test("homo_sapiens pacbio [vcf] [csi] [bam] [bai] [fasta]") {
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/vcf/NA03697B2_new.pbmm2.repeats.vcf.gz', checkIfExists: true),
+                ]
+
+                input[1] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/vcf/NA03697B2_new.pbmm2.repeats.vcf.gz.csi', checkIfExists: true),
+                ]
+
+                input[2] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/bam/NA03697B2_downsampled.pbmm2.repeats.bam', checkIfExists: true),
+                ]
+
+                input[3] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/bam/NA03697B2_downsampled.pbmm2.repeats.bam.bai', checkIfExists: true),
+                ]
+
+                input[4] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/genome3.fasta', checkIfExists: true),
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert file(process.out.vcf[0][1]).name == "test.phased.vcf" },
+                { assert file(process.out.csv[0][1]).name == "test.stats.csv" },
+                { assert snapshot(process.out).match() }
+
+            )
+        }
+
+    }
+
+    test("hiphase - vcf - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/vcf/NA03697B2_new.pbmm2.repeats.vcf.gz', checkIfExists: true),
+                ]
+
+                input[1] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/vcf/NA03697B2_new.pbmm2.repeats.vcf.gz.csi', checkIfExists: true),
+                ]
+
+                input[2] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/bam/NA03697B2_downsampled.pbmm2.repeats.bam', checkIfExists: true),
+                ]
+
+                input[3] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/bam/NA03697B2_downsampled.pbmm2.repeats.bam.bai', checkIfExists: true),
+                ]
+
+                input[4] = [
+                    [ id:'test' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/genome3.fasta', checkIfExists: true),
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}

modules/nf-core/hiphase/tests/main.nf.test.snap

@@ -0,0 +1,100 @@
+{
+    "hiphase - vcf - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.phased.vcf:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "1": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.stats.csv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "2": [
+                    "versions.yml:md5,bdbe594b7912bdb1acefa03993d4ddf7"
+                ],
+                "csv": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.stats.csv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "vcf": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.phased.vcf:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,bdbe594b7912bdb1acefa03993d4ddf7"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.1",
+            "nextflow": "24.04.4"
+        },
+        "timestamp": "2024-10-25T15:30:16.272259201"
+    },
+    "homo_sapiens pacbio [vcf] [csi] [bam] [bai] [fasta]": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.phased.vcf:md5,8d27a1ef3c4b28662a7559ea2ccc4681"
+                    ]
+                ],
+                "1": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.stats.csv:md5,9901e909f6898db2b2a9d083a4f478d5"
+                    ]
+                ],
+                "2": [
+                    "versions.yml:md5,bdbe594b7912bdb1acefa03993d4ddf7"
+                ],
+                "csv": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.stats.csv:md5,9901e909f6898db2b2a9d083a4f478d5"
+                    ]
+                ],
+                "vcf": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.phased.vcf:md5,8d27a1ef3c4b28662a7559ea2ccc4681"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,bdbe594b7912bdb1acefa03993d4ddf7"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.1",
+            "nextflow": "24.04.4"
+        },
+        "timestamp": "2024-10-29T17:04:42.688669247"
+    }
+}