Added BBMAP/REPAIR submodule (#7249)

* New submodule: BBMAP/REPAIR * BBMAP/repair snapshot added * removed cpu setting in nextflow.config --------- Co-authored-by: irongraft <[email protected]>
nf-core · Jan 7, 2025 · 7053802 · 7053802
1 parent cb8e4c9
commit 7053802
Show file tree

Hide file tree

Showing 7 changed files with 409 additions and 0 deletions.
diff --git a/modules/nf-core/bbmap/repair/environment.yml b/modules/nf-core/bbmap/repair/environment.yml
@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::bbmap=39.13"
diff --git a/modules/nf-core/bbmap/repair/main.nf b/modules/nf-core/bbmap/repair/main.nf
@@ -0,0 +1,57 @@
+process BBMAP_REPAIR {
+    tag "$meta.id"
+    label 'process_single'
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/bbmap:39.13--he5f24ec_1':
+        'biocontainers/bbmap:39.13--he5f24ec_1' }"
+
+    input:
+    tuple val(meta), path(reads)
+    val(interleave)
+
+    output:
+    tuple val(meta), path("*_repaired.fastq.gz")         , emit: repaired
+    tuple val(meta), path("${prefix}_singleton.fastq.gz"), emit: singleton
+    path  "versions.yml"                                 , emit: versions
+    path  "*.log"                                        , emit: log
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args   = task.ext.args ?: ''
+    prefix = task.ext.prefix ?: "${meta.id}"
+    in_reads  = ( interleave )  ? "in=${reads[0]}" : "in=${reads[0]} in2=${reads[1]}"
+    out_reads = ( interleave )  ? "out=${prefix}_repaired.fastq.gz outs=${prefix}_singleton.fastq.gz"
+                                : "out=${prefix}_1_repaired.fastq.gz out2=${prefix}_2_repaired.fastq.gz outs=${prefix}_singleton.fastq.gz"
+    """
+    maxmem=\$(echo \"$task.memory\"| sed 's/ GB/g/g')
+    repair.sh \\
+        -Xmx\$maxmem \\
+        $in_reads \\
+        $out_reads \\
+        threads=${task.cpus}
+        ${args} \\
+        &> ${prefix}.repair.sh.log
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        repair.sh: \$(repair.sh --version |& sed '2!d ; s/BBMap version //')
+    END_VERSIONS
+    """
+
+    stub:
+    prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    echo "" | gzip > ${prefix}_1_repaired.fastq.gz
+    echo "" | gzip > ${prefix}_2_repaired.fastq.gz
+    echo "" | gzip > ${prefix}_singleton.fastq.gz
+    touch ${prefix}.repair.sh.log
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        repair.sh: \$(repair.sh --version |& sed '2!d ; s/BBMap version //')
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/bbmap/repair/meta.yml b/modules/nf-core/bbmap/repair/meta.yml
@@ -0,0 +1,67 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "bbmap_repair"
+description: Re-pairs reads that became disordered or had some mates eliminated.
+keywords:
+  - paired reads re-pairing
+  - fastq
+  - preprocessing
+tools:
+  - repair:
+      description: Repair.sh is a tool that re-pairs reads that became disordered or had some mates eliminated
+        tools.
+      homepage: https://jgi.doe.gov/data-and-tools/software-tools/bbtools/
+      documentation: https://jgi.doe.gov/data-and-tools/software-tools/bbtools/bb-tools-user-guide/
+      licence: ["UC-LBL license (see package)"]
+      identifier: biotools:bbmap
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - reads:
+        type: file
+        description: |
+          List of input paired end fastq files
+        pattern: "*.{fastq,fq}.gz"
+  - - interleave:
+        type: boolean
+        description: |
+          Indicates whether the input paired reads are interleaved or not
+output:
+  - repaired:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*_repaired.fastq.gz":
+          type: file
+          description: re-paired reads
+          pattern: "*_repaired.fastq.gz"
+  - singleton:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "${prefix}_singleton.fastq.gz":
+          type: file
+          description: singleton reads
+          pattern: "*singleton.fastq.gz"
+  - versions:
+      - versions.yml:
+          type: file
+          description: File containing software versions
+          pattern: "versions.yml"
+  - log:
+      - "*.log":
+          type: file
+          description: log file containing stdout and stderr from repair.sh
+          pattern: "*.log"
+authors:
+  - "@mazzalab"
+maintainers:
+  - "@mazzalab"
+  - "@tm4zza"
diff --git a/modules/nf-core/bbmap/repair/tests/main.nf.test b/modules/nf-core/bbmap/repair/tests/main.nf.test
@@ -0,0 +1,115 @@
+nextflow_process {
+
+    name "Test Process BBMAP_REPAIR"
+    script "../main.nf"
+    process "BBMAP_REPAIR"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "bbmap"
+    tag "bbmap/repair"
+
+    test("sarscov2_illumina_paired - fastq_gz") {
+        config "./nextflow.config"
+
+        when {
+            params {
+                module_args = 'qin=33'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_1.fastq.gz', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_2.fastq.gz', checkIfExists: true)]
+                    ]
+                input[1] = false
+                """
+            }
+        }
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out.versions).match() }
+            )
+        }
+    }
+
+    test("sarscov2_illumina_interleaved - fastq_gz") {
+        config "./nextflow.config"
+
+        when {
+            params {
+                module_args = 'qin=33'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_interleaved.fastq.gz', checkIfExists: true)]
+                    ]
+                input[1] = true
+                """
+            }
+        }
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out.versions).match() }
+            )
+        }
+    }
+
+    test("sarscov2_illumina_paired - fastq_gz - stub") {
+        config "./nextflow.config"
+        options "-stub"
+
+        when {
+            params {
+                module_args = 'qin=33'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_1.fastq.gz', checkIfExists: true),
+                    file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_2.fastq.gz', checkIfExists: true)]
+                    ]
+                input[1] = false
+                """
+            }
+        }
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+
+    test("sarscov2_illumina_interleaved - fastq_gz - stub") {
+        config "./nextflow.config"
+        options "-stub"
+
+        when {
+            params {
+                module_args = 'qin=33'
+            }
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [file(params.modules_testdata_base_path + 'genomics/sarscov2/illumina/fastq/test_interleaved.fastq.gz', checkIfExists: true)]
+                    ]
+                input[1] = true
+                """
+            }
+        }
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+}