Add gapseq modules for bacterial metabolic pathway analysis

modules/nf-core/gapseq/doall/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::gapseq=1.4.0

modules/nf-core/gapseq/doall/main.nf

@@ -0,0 +1,48 @@
+process GAPSEQ_DOALL {
+    tag "${meta.id}"
+    label 'process_high'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gapseq:1.4.0--hdfd78af_0' :
+        'biocontainers/gapseq:1.4.0--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(fasta), path(medium)
+
+    output:
+    tuple val(meta), path("*.RDS")  , emit: model
+    tuple val(meta), path("*.tbl")  , emit: tbl
+    tuple val(meta), path("*.fna")  , emit: fna      , optional: true
+    tuple val(meta), path("*.log")  , emit: log      , optional: true
+    path  "versions.yml"            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def medium_arg = medium ? "-m $medium" : ''
+    """
+    gapseq \\
+        doall \\
+        -p ${task.cpus} \\
+        $medium_arg \\
+        $args \\
+        $fasta
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gapseq: \$(gapseq --version 2>&1 | sed 's/gapseq //')
+    END_VERSIONS
+    """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}_model-filled.RDS
+    touch ${prefix}_pathways.tbl
+    touch ${prefix}_transporters.tbl
+    """
+}

modules/nf-core/gapseq/doall/meta.yml

@@ -0,0 +1,92 @@
+name: gapseq_doall
+description: Complete gapseq workflow from genome to gap-filled model
+keywords:
+  - metabolic model
+  - gap filling
+  - pathway prediction
+  - bacteria
+  - genome-scale model
+  - complete workflow
+tools:
+  - gapseq:
+      description: |
+        gapseq is designed to combine metabolic pathway analysis with metabolic network reconstruction and curation.
+        The doall command runs the complete workflow: pathway prediction, transporter inference, model construction, and gap filling.
+      homepage: https://github.com/jotech/gapseq
+      documentation: https://gapseq.readthedocs.io/
+      tool_dev_url: https://github.com/jotech/gapseq
+      doi: "10.1186/s13059-021-02295-1"
+      licence: ["AGPL-3.0-only"]
+      identifier: ""
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. [ id:'test', single_end:false ]
+    - fasta:
+        type: file
+        description: Protein or genomic FASTA file
+        pattern: "*.{fasta,fa,faa,fna,gz}"
+        ontologies: []
+    - medium:
+        type: file
+        description: Optional medium definition file (CSV format)
+        pattern: "*.csv"
+        ontologies: []
+output:
+  model:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.RDS":
+          type: file
+          description: Gap-filled metabolic model in RDS format
+          pattern: "*.RDS"
+          ontologies: []
+  tbl:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.tbl":
+          type: file
+          description: TBL files containing pathway and transporter predictions
+          pattern: "*.tbl"
+          ontologies: []
+  fna:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.fna":
+          type: file
+          description: Optional extracted sequences in FASTA format
+          pattern: "*.fna"
+          ontologies: []
+  log:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.log":
+          type: file
+          description: Log files from gapseq doall
+          pattern: "*.log"
+          ontologies: []
+  versions:
+    - versions.yml:
+        type: file
+        description: File containing software versions
+        pattern: "versions.yml"
+        ontologies:
+          - edam: http://edamontology.org/format_3750 # YAML
+authors:
+  - "@copilot"
+maintainers:
+  - "@copilot"

modules/nf-core/gapseq/doall/tests/main.nf.test

@@ -0,0 +1,57 @@
+nextflow_process {
+
+    name "Test Process GAPSEQ_DOALL"
+    script "../main.nf"
+    process "GAPSEQ_DOALL"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "gapseq"
+    tag "gapseq/doall"
+
+    test("sarscov2 genome [fasta]") {
+
+        when {
+            process {
+                """
+                input[0] = Channel.of([
+                    [ id: 'test', single_end:false ],
+                    [ file(params.modules_testdata_base_path + 'genomics/sarscov2/genome/genome.fasta', checkIfExists: true) ],
+                    []
+                ])
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+
+    test("sarscov2 genome [fasta] - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = Channel.of([
+                    [ id: 'test', single_end:false ],
+                    [ file(params.modules_testdata_base_path + 'genomics/sarscov2/genome/genome.fasta', checkIfExists: true) ],
+                    []
+                ])
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+}

modules/nf-core/gapseq/draft/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::gapseq=1.4.0

modules/nf-core/gapseq/draft/main.nf

@@ -0,0 +1,44 @@
+process GAPSEQ_DRAFT {
+    tag "${meta.id}"
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/gapseq:1.4.0--hdfd78af_0' :
+        'biocontainers/gapseq:1.4.0--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(pathways), path(transporters)
+
+    output:
+    tuple val(meta), path("*.RDS")  , emit: draft
+    tuple val(meta), path("*.log")  , emit: log      , optional: true
+    path  "versions.yml"            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def transporters_arg = transporters ? "-t $transporters" : ''
+    """
+    gapseq \\
+        draft \\
+        -r $pathways \\
+        $transporters_arg \\
+        -c $prefix \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gapseq: \$(gapseq --version 2>&1 | sed 's/gapseq //')
+    END_VERSIONS
+    """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}-draft.RDS
+    """
+}

modules/nf-core/gapseq/draft/meta.yml

@@ -0,0 +1,68 @@
+name: gapseq_draft
+description: Create draft metabolic model from pathway and transporter predictions
+keywords:
+  - metabolic model
+  - draft reconstruction
+  - bacteria
+  - genome-scale model
+tools:
+  - gapseq:
+      description: |
+        gapseq is designed to combine metabolic pathway analysis with metabolic network reconstruction and curation.
+        The draft command constructs a draft metabolic model from pathway and transporter predictions.
+      homepage: https://github.com/jotech/gapseq
+      documentation: https://gapseq.readthedocs.io/
+      tool_dev_url: https://github.com/jotech/gapseq
+      doi: "10.1186/s13059-021-02295-1"
+      licence: ["AGPL-3.0-only"]
+      identifier: ""
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. [ id:'test', single_end:false ]
+    - pathways:
+        type: file
+        description: Pathways CSV file from gapseq find
+        pattern: "*-all-Pathways.tbl"
+        ontologies: []
+    - transporters:
+        type: file
+        description: Transporters CSV file from gapseq find-transport
+        pattern: "*-Transporter.tbl"
+        ontologies: []
+output:
+  draft:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.RDS":
+          type: file
+          description: Draft metabolic model in RDS format
+          pattern: "*.RDS"
+          ontologies: []
+  log:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. [ id:'test', single_end:false ]
+      - "*.log":
+          type: file
+          description: Log files from gapseq draft
+          pattern: "*.log"
+          ontologies: []
+  versions:
+    - versions.yml:
+        type: file
+        description: File containing software versions
+        pattern: "versions.yml"
+        ontologies:
+          - edam: http://edamontology.org/format_3750 # YAML
+authors:
+  - "@copilot"
+maintainers:
+  - "@copilot"

modules/nf-core/gapseq/draft/tests/main.nf.test

@@ -0,0 +1,35 @@
+nextflow_process {
+
+    name "Test Process GAPSEQ_DRAFT"
+    script "../main.nf"
+    process "GAPSEQ_DRAFT"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "gapseq"
+    tag "gapseq/draft"
+
+    test("test draft model creation - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = Channel.of([
+                    [ id: 'test' ],
+                    file('test-all-Pathways.tbl'),
+                    file('test-Transporter.tbl')
+                ])
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+    }
+}

modules/nf-core/gapseq/fill/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::gapseq=1.4.0