add custom/splitfastqbylane module

modules/msk/custom/splitfastqbylane/environment.yml

@@ -0,0 +1,9 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+name: "custom_splitfastqbylane"
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - "anaconda::gawk=5.1.0"

modules/msk/custom/splitfastqbylane/main.nf

@@ -0,0 +1,45 @@
+process CUSTOM_SPLITFASTQBYLANE {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gawk:5.1.0':
+        'biocontainers/gawk:5.1.0' }"
+
+    input:
+    tuple val(meta), path(reads)
+
+    output:
+    tuple val(meta), path("*.split.fastq.gz"), emit: reads
+    path "versions.yml"                      , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def read1 = [reads].flatten()[0]
+    def read2 = [reads].flatten().size() > 1 ? reads[1] : '' 
+    """
+    split_lanes_awk.sh ${prefix} ${read1} ${read2}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gawk: \$(awk -Wversion | sed '1!d; s/.*Awk //; s/,.*//')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch out.split.fastq
+    gzip out.split.fastq
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gawk: \$(awk -Wversion | sed '1!d; s/.*Awk //; s/,.*//')
+    END_VERSIONS
+    """
+}

modules/msk/custom/splitfastqbylane/meta.yml

@@ -0,0 +1,46 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "custom_splitfastqbylane"
+description: Split fastq into multiple fastqs by lane
+keywords:
+  - awk
+  - fastq
+  - split
+tools:
+  - "custom":
+      description: "GNU awk"
+      homepage: "https://www.gnu.org/software/gawk/manual/gawk.html"
+      licence: ["GPL v3"]
+
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. `[ id:'sample1', single_end:false ]`
+
+  - reads:
+      type: file
+      description: Paired end or single end FASTQ file(s)
+      pattern: "*.{fastq,fastq.gz}"
+
+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"
+  - reads:
+      type: file
+      description: Output fastq files containing only one read lane per file.
+      pattern: "*.{split.fastq.gz}"
+
+authors:
+  - "@anoronh4"
+maintainers:
+  - "@anoronh4"

modules/msk/custom/splitfastqbylane/resources/usr/bin/split_lanes_awk.sh

@@ -0,0 +1,38 @@
+#!/bin/bash
+
+prefix=$1
+read1=$2
+read2=$3
+
+
+if [[ "${read1}" == *gz ]] ; then
+    cat_="zcat"
+else
+    cat_="cat"
+fi
+
+function a() {
+    awk \
+        -v prefix=$3 \
+        -v readnumber=$1 \
+        '
+        BEGIN {FS = ":"}
+        {
+            lane=$(NF-3)
+            flowcell=$(NF-4)
+            outfastq=prefix"@"flowcell"_L00"lane"_R"readnumber".split.fastq.gz"
+            print | "gzip > "outfastq
+            for (i = 1; i <= 3; i++) {
+                getline
+                print | "gzip > "outfastq
+            }
+        }
+        ' <( eval "$cat_ $2")
+}
+
+echo "processing read1"
+a 1 ${read1} ${prefix}
+if [ ! -z ${read2} ] ; then
+    echo "processing read2"
+    a 2 ${read2} ${prefix}
+fi

modules/msk/custom/splitfastqbylane/tests/main.nf.test

@@ -0,0 +1,64 @@
+nextflow_process {
+
+    name "Test Process CUSTOM_SPLITFASTQBYLANE"
+    script "../main.nf"
+    process "CUSTOM_SPLITFASTQBYLANE"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "custom"
+    tag "custom/splitfastqbylane"
+
+    test("homo_sapiens - fastq") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [
+                        file(params.test_data['sarscov2']['illumina']['test_1_fastq_gz'], checkIfExists: true),
+                        file(params.test_data['sarscov2']['illumina']['test_2_fastq_gz'], checkIfExists: true)
+                    ]
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+    test("sarscov2 - fastq - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    [
+                        file(params.test_data['sarscov2']['illumina']['test_1_fastq_gz'], checkIfExists: true),
+                        file(params.test_data['sarscov2']['illumina']['test_2_fastq_gz'], checkIfExists: true)
+                    ]
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}

modules/msk/custom/splitfastqbylane/tests/main.nf.test.snap

@@ -0,0 +1,82 @@
+{
+    "sarscov2 - fastq - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "out.split.fastq.gz:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,68b779fcdaf62c97ffcd67d49ab59b0e"
+                ],
+                "reads": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "out.split.fastq.gz:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,68b779fcdaf62c97ffcd67d49ab59b0e"
+                ]
+            }
+        ],
+        "timestamp": "2024-01-16T11:28:05.672874"
+    },
+    "homo_sapiens - fastq": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        [
+                            "test@HK3MMAFX2_L001_R1.split.fastq.gz:md5,6fdae9563e6f3428ea953435cb71f762",
+                            "test@HK3MMAFX2_L001_R2.split.fastq.gz:md5,a2315ba81e2993823e38af0b83b9e7bd",
+                            "test@HK3MMAFX2_L002_R1.split.fastq.gz:md5,3c5e1d1bf430c964fa2ac1b7faf8c313",
+                            "test@HK3MMAFX2_L002_R2.split.fastq.gz:md5,f3aee4f14f59bd2e26753ec0dda1d5dc",
+                            "test@HK3MMAFX2_L003_R1.split.fastq.gz:md5,c98453d33b9bb4ff224ae5571489913b",
+                            "test@HK3MMAFX2_L003_R2.split.fastq.gz:md5,898a6d549be78b5214c3d5647e7c3c49",
+                            "test@HK3MMAFX2_L004_R1.split.fastq.gz:md5,11b776b50ea9bcc6670aaca952f6a939",
+                            "test@HK3MMAFX2_L004_R2.split.fastq.gz:md5,a50125ecb428a7f534a944c5a22c8109"
+                        ]
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,68b779fcdaf62c97ffcd67d49ab59b0e"
+                ],
+                "reads": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        [
+                            "test@HK3MMAFX2_L001_R1.split.fastq.gz:md5,6fdae9563e6f3428ea953435cb71f762",
+                            "test@HK3MMAFX2_L001_R2.split.fastq.gz:md5,a2315ba81e2993823e38af0b83b9e7bd",
+                            "test@HK3MMAFX2_L002_R1.split.fastq.gz:md5,3c5e1d1bf430c964fa2ac1b7faf8c313",
+                            "test@HK3MMAFX2_L002_R2.split.fastq.gz:md5,f3aee4f14f59bd2e26753ec0dda1d5dc",
+                            "test@HK3MMAFX2_L003_R1.split.fastq.gz:md5,c98453d33b9bb4ff224ae5571489913b",
+                            "test@HK3MMAFX2_L003_R2.split.fastq.gz:md5,898a6d549be78b5214c3d5647e7c3c49",
+                            "test@HK3MMAFX2_L004_R1.split.fastq.gz:md5,11b776b50ea9bcc6670aaca952f6a939",
+                            "test@HK3MMAFX2_L004_R2.split.fastq.gz:md5,a50125ecb428a7f534a944c5a22c8109"
+                        ]
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,68b779fcdaf62c97ffcd67d49ab59b0e"
+                ]
+            }
+        ],
+        "timestamp": "2024-01-16T11:27:57.305806"
+    }
+}

modules/msk/custom/splitfastqbylane/tests/tags.yml

@@ -0,0 +1,2 @@
+custom/splitfastqbylane:
+  - "modules/msk/custom/splitfastqbylane/**"

tests/config/nf-test.config

@@ -46,3 +46,6 @@ includeConfig 'test_data.config'
 manifest {
     nextflowVersion = '!>=23.04.0'
 }
+
+nextflow.enable.moduleBinaries = true
+