Add FCSGX_CLEANGENOME (#6975)

Co-authored-by: Lauren huet <[email protected]> Co-authored-by: Sateesh_Peri <[email protected]> Co-authored-by: Edmund Miller <[email protected]>
nf-core · Nov 14, 2024 · b48e64d · b48e64d
1 parent 4265ef4
commit b48e64d
Show file tree

Hide file tree

Showing 5 changed files with 307 additions and 0 deletions.
diff --git a/modules/nf-core/fcsgx/cleangenome/environment.yml b/modules/nf-core/fcsgx/cleangenome/environment.yml
@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::ncbi-fcs-gx=0.5.4"
diff --git a/modules/nf-core/fcsgx/cleangenome/main.nf b/modules/nf-core/fcsgx/cleangenome/main.nf
@@ -0,0 +1,51 @@
+process FCSGX_CLEANGENOME {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/ncbi-fcs-gx:0.5.4--h4ac6f70_1':
+        'biocontainers/ncbi-fcs-gx:0.5.4--h4ac6f70_1' }"
+
+    input:
+    tuple val(meta), path(fasta), path(fcsgx_report)
+
+    output:
+    tuple val(meta), path("*.cleaned.fasta")     , emit: cleaned
+    tuple val(meta), path("*.contaminants.fasta"), emit: contaminants
+    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}"
+    """
+    gx \\
+        clean-genome \\
+        --input ${fasta} \\
+        --action-report ${fcsgx_report} \\
+        --output ${prefix}.cleaned.fasta \\
+        --contam-fasta-out ${prefix}.contaminants.fasta \\
+        ${args}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        fcsgx: \$( gx --help | sed '/build/!d; s/.*:v//; s/-.*//' )
+    END_VERSIONS
+    """
+
+    stub:
+    // def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.cleaned.fasta
+    touch ${prefix}.contaminants.fasta
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        fcsgx: \$( gx --help | sed '/build/!d; s/.*:v//; s/-.*//' )
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/fcsgx/cleangenome/meta.yml b/modules/nf-core/fcsgx/cleangenome/meta.yml
@@ -0,0 +1,64 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "fcsgx_cleangenome"
+description: Runs FCS-GX (Foreign Contamination Screen - Genome eXtractor) to
+  remove foreign contamination from genome assemblies
+keywords:
+  - genome
+  - assembly
+  - contamination
+  - screening
+  - cleaning
+  - fcs-gx
+tools:
+  - "fcsgx":
+      description: "The NCBI Foreign Contamination Screen. Genomic cross-species aligner,
+        for contamination detection."
+      homepage: "https://github.com/ncbi/fcs-gx"
+      documentation: "https://github.com/ncbi/fcs/wiki/"
+      tool_dev_url: "https://github.com/ncbi/fcs-gx"
+      doi: "10.1186/s13059-024-03198-7"
+      licence: ["NCBI-PD"]
+      identifier: "biotools:ncbi_fcs"
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - fasta:
+        type: file
+        description: Input genome assembly file in FASTA format
+        pattern: "*.{fa,fasta,fna}"
+    - fcsgx_report:
+        type: file
+        description: Final contamination report with contaminant cleaning actions. Generated using FCSGX_RUNGX
+output:
+  - cleaned:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.cleaned.fasta":
+          type: file
+          description: The fasta file after cleaning, where sequences annotated as ACTION_EXCLUDE or ACTION_TRIM are excluded
+  - contaminants:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.contaminants.fasta":
+          type: file
+          description: Sequences annotated as ACTION_EXCLUDE which are marked as contaminants.
+  - versions:
+      - versions.yml:
+          type: file
+          description: File containing software versions
+          pattern: "versions.yml"
+authors:
+  - "@mahesh-panchal"
+  - "@LaurenHuet"
+maintainers:
+  - "@mahesh-panchal"
diff --git a/modules/nf-core/fcsgx/cleangenome/tests/main.nf.test b/modules/nf-core/fcsgx/cleangenome/tests/main.nf.test
@@ -0,0 +1,85 @@
+nextflow_process {
+
+    name "Test Process FCSGX_CLEANGENOME"
+    script "../main.nf"
+    process "FCSGX_CLEANGENOME"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "fcsgx"
+    tag "fcsgx/fetchdb"
+    tag "fcsgx/rungx"
+    tag "fcsgx/cleangenome"
+
+    setup {
+        run("FCSGX_FETCHDB"){
+            script "../../fetchdb/main.nf"
+            process {
+                """
+                input[0] = file('https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.manifest', checkIfExists: true)
+                """
+            }
+        }
+        run("FCSGX_RUNGX") {
+            script "../../rungx/main.nf"
+            process {
+                """
+                input[0] = [
+                    [ id:'test' ], // meta map
+                    '2697049', // taxid for SARS-CoV-2
+                    file('https://zenodo.org/records/10932013/files/FCS_combo_test.fa', checkIfExists: true),
+                ]
+                input[1] = FCSGX_FETCHDB.out.database
+                input[2] = []
+                """
+            }
+        }
+    }
+
+    test("sarscov2 - fasta") {
+
+        when {
+            process {
+                """
+                input[0] = Channel.value([
+                    [ id:'test' ], // meta map
+                    file('https://zenodo.org/records/10932013/files/FCS_combo_test.fa', checkIfExists: true),
+                ]).join(FCSGX_RUNGX.out.fcsgx_report)
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+    test("sarscov2 - fasta - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = Channel.value([
+                    [ id:'test' ], // meta map
+                    file('https://zenodo.org/records/10932013/files/FCS_combo_test.fa', checkIfExists: true),
+                ]).join(FCSGX_RUNGX.out.fcsgx_report)
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}
diff --git a/modules/nf-core/fcsgx/cleangenome/tests/main.nf.test.snap b/modules/nf-core/fcsgx/cleangenome/tests/main.nf.test.snap
@@ -0,0 +1,100 @@
+{
+    "sarscov2 - fasta - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.cleaned.fasta:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "1": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.contaminants.fasta:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "2": [
+                    "versions.yml:md5,4235d280269123cbd3e43d920bcf71b9"
+                ],
+                "cleaned": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.cleaned.fasta:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "contaminants": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.contaminants.fasta:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,4235d280269123cbd3e43d920bcf71b9"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.2",
+            "nextflow": "24.10.0"
+        },
+        "timestamp": "2024-11-12T10:02:26.605398704"
+    },
+    "sarscov2 - fasta": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.cleaned.fasta:md5,59ba4b4c773f02a0dc90034762aa1ddf"
+                    ]
+                ],
+                "1": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.contaminants.fasta:md5,17c11ea7d2d0076be5183b87618ee91e"
+                    ]
+                ],
+                "2": [
+                    "versions.yml:md5,4235d280269123cbd3e43d920bcf71b9"
+                ],
+                "cleaned": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.cleaned.fasta:md5,59ba4b4c773f02a0dc90034762aa1ddf"
+                    ]
+                ],
+                "contaminants": [
+                    [
+                        {
+                            "id": "test"
+                        },
+                        "test.contaminants.fasta:md5,17c11ea7d2d0076be5183b87618ee91e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,4235d280269123cbd3e43d920bcf71b9"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.2",
+            "nextflow": "24.10.0"
+        },
+        "timestamp": "2024-11-12T12:05:43.065207163"
+    }
+}