Add module and subworkflow mirtop (#6587)

* add mirtop gff module * add mirtop/counts module * add mirtop/export module * add mirtop/stats module * update yml * add bam_stats_mirna_mirtop subworkflow * add required dependencies * update test * update subworkflow test * update yml and output files * fix tag linting issue? * fix linting * run prettier * try to solve conda issue * add a compatible pandas * fix linting * update failing ci mirtop/gff test
nf-core · Sep 6, 2024 · 9897e69 · 9897e69
1 parent c3e4418
commit 9897e69
Show file tree

Hide file tree

Showing 26 changed files with 1,344 additions and 0 deletions.
diff --git a/modules/nf-core/mirtop/counts/environment.yml b/modules/nf-core/mirtop/counts/environment.yml
@@ -0,0 +1,9 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::mirtop=0.4.25"
+  - "conda-forge::r-base=4.1.1"
+  - "conda-forge::r-data.table=1.14.2"
diff --git a/modules/nf-core/mirtop/counts/main.nf b/modules/nf-core/mirtop/counts/main.nf
@@ -0,0 +1,53 @@
+process MIRTOP_COUNTS {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/mulled-v2-0c13ef770dd7cc5c76c2ce23ba6669234cf03385:63be019f50581cc5dfe4fc0f73ae50f2d4d661f7-0' :
+        'biocontainers/mulled-v2-0c13ef770dd7cc5c76c2ce23ba6669234cf03385:63be019f50581cc5dfe4fc0f73ae50f2d4d661f7-0' }"
+
+    input:
+    tuple val(meta), path(mirtop_gff)
+    tuple val(meta2), path(hairpin)
+    tuple val(meta3), path(gtf), val(species)
+
+    output:
+    tuple val(meta), path("counts/*.tsv"), emit: tsv
+    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}"
+    """
+    mirtop \\
+        counts \\
+        $args \\
+        --hairpin $hairpin \\
+        --gtf $gtf \\
+        --sps $species \\
+        --gff $mirtop_gff \\
+        -o counts
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        mirtop: \$(echo \$(mirtop --version 2>&1) | sed 's/^.*mirtop //')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    mkdir counts
+    touch counts/mirtop.tsv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        mirtop: \$(echo \$(mirtop --version 2>&1) | sed 's/^.*mirtop //')
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/mirtop/counts/meta.yml b/modules/nf-core/mirtop/counts/meta.yml
@@ -0,0 +1,57 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "mirtop_counts"
+description: mirtop counts generates a file with the minimal information about each sequence and the count data in columns for each samples.
+keywords:
+  - mirna
+  - isomir
+  - gff
+tools:
+  - "mirtop":
+      description: "Small RNA-seq annotation"
+      homepage: "https://github.com/miRTop/mirtop"
+      documentation: "https://mirtop.readthedocs.io/en/latest/"
+      tool_dev_url: "https://github.com/miRTop/mirtop"
+      licence: ["MIT License"]
+
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. `[ id:'sample1', single_end:false ]`
+  - mirtop_gff:
+      type: file
+      description: GFF file
+      pattern: "*.{gff}"
+  - hairpin:
+      type: file
+      description: Hairpin file
+      pattern: "*.{fa,fasta}"
+  - gtf:
+      type: file
+      description: GTF file
+      pattern: "*.{gtf}"
+  - species:
+      type: string
+      description: Species name of the GTF file
+
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. `[ id:'sample1', single_end:false ]`
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+  - tsv:
+      type: file
+      description: TSV file
+      pattern: "*.{tsv}"
+
+authors:
+  - "@atrigila"
+maintainers:
+  - "@atrigila"
diff --git a/modules/nf-core/mirtop/counts/tests/main.nf.test b/modules/nf-core/mirtop/counts/tests/main.nf.test
@@ -0,0 +1,97 @@
+nextflow_process {
+
+    name "Test Process MIRTOP_COUNTS"
+    script "../main.nf"
+    config "./nextflow.config"
+    process "MIRTOP_COUNTS"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "mirtop"
+    tag "mirtop/gff"
+    tag "mirtop/counts"
+
+    setup {
+            run("MIRTOP_GFF") {
+                script "../../gff/main.nf"
+                process {
+                """
+                input[0] = [
+                    [ id:'sample_sim_isomir_bam'], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/sim_isomir_sort.bam", checkIfExists: true),
+                ]
+                input[1] = [
+                    [ id:'hairpin'], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hairpin_mirtop.fa", checkIfExists: true),
+                ]
+                input[2] = [
+                    [ id:'hsa' ], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hsa.gff3", checkIfExists: true),
+                "hsa"]
+                """
+            }
+            }
+        }
+
+    test("isomir - bam") {
+
+        when {
+            process {
+                """
+                input[0] = MIRTOP_GFF.out.mirtop_gff
+                input[1] = [
+                    [ id:'hairpin'], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hairpin_mirtop.fa", checkIfExists: true),
+                ]
+                input[2] = [
+                    [ id:'hsa' ], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hsa.gff3", checkIfExists: true),
+                "hsa"]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    process.out.versions
+                        ).match()
+                        },
+                // md5sum is not stable, order of elements in column "variant" change
+                { assert file(process.out.tsv[0][1]).readLines().findAll { it.contains("iso-22-I0S21NSLN") }}
+            )
+        }
+
+    }
+
+    test("isomir - bam - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = MIRTOP_GFF.out.mirtop_gff
+                input[1] = [
+                    [ id:'hairpin'], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hairpin_mirtop.fa", checkIfExists: true),
+                ]
+                input[2] = [
+                    [ id:'hsa' ], // meta map
+                    file("https://github.com/nf-core/test-datasets/raw/modules/data/delete_me/mirtop/hsa.gff3", checkIfExists: true),
+                "hsa"]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}
diff --git a/modules/nf-core/mirtop/counts/tests/main.nf.test.snap b/modules/nf-core/mirtop/counts/tests/main.nf.test.snap
@@ -0,0 +1,47 @@
+{
+    "isomir - bam - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "sample_sim_isomir_bam"
+                        },
+                        "mirtop.tsv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,16430682c2080d67387e7d5e9caf69a7"
+                ],
+                "tsv": [
+                    [
+                        {
+                            "id": "sample_sim_isomir_bam"
+                        },
+                        "mirtop.tsv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,16430682c2080d67387e7d5e9caf69a7"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.0",
+            "nextflow": "24.04.4"
+        },
+        "timestamp": "2024-09-03T13:24:41.148536938"
+    },
+    "isomir - bam": {
+        "content": [
+            [
+                "versions.yml:md5,16430682c2080d67387e7d5e9caf69a7"
+            ]
+        ],
+        "meta": {
+            "nf-test": "0.9.0",
+            "nextflow": "24.04.4"
+        },
+        "timestamp": "2024-09-04T18:39:50.607483472"
+    }
+}
diff --git a/modules/nf-core/mirtop/counts/tests/nextflow.config b/modules/nf-core/mirtop/counts/tests/nextflow.config
@@ -0,0 +1,5 @@
+process {
+    withName: 'MIRTOP_COUNTS' {
+        ext.args = '--add-extra'
+    }
+}
diff --git a/modules/nf-core/mirtop/export/environment.yml b/modules/nf-core/mirtop/export/environment.yml
@@ -0,0 +1,10 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::mirtop=0.4.25"
+  - "bioconda::samtools=1.15.1"
+  - "conda-forge::r-base=4.1.1"
+  - "conda-forge::r-data.table=1.14.2"
diff --git a/modules/nf-core/mirtop/export/main.nf b/modules/nf-core/mirtop/export/main.nf
@@ -0,0 +1,57 @@
+process MIRTOP_EXPORT {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/mulled-v2-0c13ef770dd7cc5c76c2ce23ba6669234cf03385:63be019f50581cc5dfe4fc0f73ae50f2d4d661f7-0' :
+        'biocontainers/mulled-v2-0c13ef770dd7cc5c76c2ce23ba6669234cf03385:63be019f50581cc5dfe4fc0f73ae50f2d4d661f7-0' }"
+
+    input:
+    tuple val(meta), path(mirtop_gff)
+    tuple val(meta2), path(hairpin)
+    tuple val(meta3), path(gtf), val(species)
+
+    output:
+    tuple val(meta), path("export/*_rawData.tsv")     , emit: tsv, optional: true
+    tuple val(meta), path("export/*.fasta")           , emit: fasta, optional: true
+    tuple val(meta), path("export/*.vcf*")            , emit: vcf  , optional: true
+    path "versions.yml"                               , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: '--format isomir'
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    mirtop \\
+        export \\
+        $args \\
+        --hairpin $hairpin\\
+        --gtf $gtf \\
+        --sps $species \\
+        -o export \\
+        $mirtop_gff
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        mirtop: \$(echo \$(mirtop --version 2>&1) | sed 's/^.*mirtop //')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    mkdir export
+    touch export/${prefix}.fasta
+    touch export/${prefix}.vcf
+    touch export/${prefix}.tsv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        mirtop: \$(echo \$(mirtop --version 2>&1) | sed 's/^.*mirtop //')
+    END_VERSIONS
+    """
+}