Merge pull request #52 from EBI-Metagenomics/new-module--taxonkit/ref…

…ormat

New module taxonkit/reformat

modules/ebi-metagenomics/taxonkit/reformat/environment.yml

@@ -0,0 +1,6 @@
+---
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::taxonkit=0.17.0"

modules/ebi-metagenomics/taxonkit/reformat/main.nf

@@ -0,0 +1,52 @@
+process TAXONKIT_REFORMAT {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/taxonkit:0.17.0--h9ee0642_1':
+        'biocontainers/taxonkit:0.17.0--h9ee0642_1' }"
+
+    input:
+    tuple val(meta), path(tsv)
+    path taxdb
+
+    output:
+    tuple val(meta), path("*.tsv"), emit: reformat_tsv
+    path "versions.yml"           , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    """
+    taxonkit \\
+        reformat \\
+        $args \\
+        --threads $task.cpus \\
+        --data-dir $taxdb \\
+        --out-file ${prefix}.tsv \\
+        $tsv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        taxonkit: \$( taxonkit version | sed 's/.* v//' )
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    """
+    touch ${prefix}.tsv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        taxonkit: \$( taxonkit version | sed 's/.* v//' )
+    END_VERSIONS
+    """
+}

modules/ebi-metagenomics/taxonkit/reformat/meta.yml

@@ -0,0 +1,51 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "taxonkit_reformat"
+description: Reformat lineage in canonical ranks
+keywords:
+  - taxonomy
+  - taxids
+  - ncbi
+  - lineage
+tools:
+  - "taxonkit":
+      description: "A Cross-platform and Efficient NCBI Taxonomy Toolkit"
+      homepage: "https://bioinf.shenwei.me/taxonkit/"
+      documentation: "https://bioinf.shenwei.me/taxonkit/usage/#reformat"
+      tool_dev_url: "https://github.com/shenwei356/taxonkit"
+      doi: "10.1016/j.jgg.2021.03.006"
+      licence: ["MIT"]
+      identifier: biotools:taxonkit
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+    - tsv:
+        type: file
+        description: Input TSV file
+        pattern: "*.{tsv,tsv.gz}"
+  - - taxdb:
+        type: file
+        description: Taxonomy database unpacked from ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz
+output:
+  - reformat_tsv:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.tsv":
+          type: file
+          description: Reformated output TSV file
+          pattern: "*.{tsv}"
+  - versions:
+      - versions.yml:
+          type: file
+          description: File containing software versions
+          pattern: "versions.yml"
+authors:
+  - "@arajkovic"
+maintainers:
+  - "@arajkovic"

modules/ebi-metagenomics/taxonkit/reformat/tests/fixtures/nucl_gb.accession2taxid

@@ -0,0 +1,3 @@
+accession	accession.version	taxid	gi
+MT192765	MT192765.1	2697049	1821109001
+NZ_LS483480	NZ_LS483480.1	727	1409087034

modules/ebi-metagenomics/taxonkit/reformat/tests/fixtures/taxdump/names.dmp

@@ -0,0 +1,26 @@
+2697049	|	2019-nCoV	|		|	equivalent name	|
+2697049	|	COVID-19 virus	|		|	equivalent name	|
+2697049	|	HCoV-19	|		|	equivalent name	|
+2697049	|	Human coronavirus 2019	|		|	equivalent name	|
+2697049	|	SARS-2	|		|	equivalent name	|
+2697049	|	SARS2	|		|	equivalent name	|
+2697049	|	SARS-CoV-2	|		|	acronym	|
+2697049	|	SARS-CoV2	|		|	equivalent name	|
+2697049	|	Severe acute respiratory syndrome coronavirus 2	|		|	scientific name	|
+727	|	ATCC 33391	|	ATCC 33391 <type strain>	|	type material	|
+727	|	"Bacterium influenzae" Lehmann and Neumann 1896	|		|	authority	|
+727	|	Bacterium influenzae	|		|	synonym	|
+727	|	CCUG 23945	|	CCUG 23945 <type strain>	|	type material	|
+727	|	CIP 102514	|	CIP 102514 <type strain>	|	type material	|
+727	|	"Coccobacillus pfeifferi" Neveu-Lemaire 1921	|		|	authority	|
+727	|	Coccobacillus pfeifferi	|		|	synonym	|
+727	|	DSM 4690	|	DSM 4690 <type strain>	|	type material	|
+727	|	Haemophilus influenzae (Lehmann and Neumann 1896) Winslow et al. 1917	|		|	authority	|
+727	|	Haemophilus influenzae	|		|	scientific name	|
+727	|	"Haemophilus meningitidis" (Martins) Hauduroy et al. 1937	|		|	authority	|
+727	|	Haemophilus meningitidis	|		|	synonym	|
+727	|	"Influenza-bacillus" Pfeiffer 1892	|		|	authority	|
+727	|	Influenza-bacillus	|		|	synonym	|
+727	|	"Mycobacterium influenzae" (Lehmann and Neumann 1896) Chester 1901	|		|	authority	|
+727	|	Mycobacterium influenzae	|		|	synonym	|
+727	|	NCTC 8143	|	NCTC 8143 <type strain>	|	type material	|

modules/ebi-metagenomics/taxonkit/reformat/tests/fixtures/taxdump/nodes.dmp

@@ -0,0 +1,22 @@
+1	|	1	|	no rank	|		|	8	|	0	|	1	|	0	|	0	|	0	|	0	|	0	|		|
+10239	|	1	|	superkingdom	|		|	9	|	0	|	1	|	0	|	0	|	0	|	0	|	0	|		|
+2559587	|	10239	|	clade	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|		|
+2732396	|	2559587	|	kingdom	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|		|
+2732408	|	2732396	|	phylum	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|		|
+2732506	|	2732408	|	class	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|		|
+76804	|	2732506	|	order	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+2499399	|	76804	|	suborder	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+11118	|	2499399	|	family	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+2501931	|	11118	|	subfamily	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+694002	|	2501931	|	genus	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+2509511	|	694002	|	subgenus	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+694009	|	2509511	|	species	|	SA	|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|	code compliant; specified	|
+2697049	|	694009	|	no rank	|		|	9	|	1	|	1	|	1	|	0	|	1	|	0	|	0	|		|
+131567	|	1	|	no rank	|		|	8	|	1	|	1	|	1	|	0	|	1	|	1	|	0	|		|
+2	|	131567	|	superkingdom	|		|	0	|	0	|	11	|	0	|	0	|	0	|	0	|	0	|		|
+1224	|	2	|	phylum	|		|	0	|	1	|	11	|	1	|	0	|	1	|	0	|	0	|		|
+1236	|	1224	|	class	|		|	0	|	1	|	11	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+135625	|	1236	|	order	|		|	0	|	1	|	11	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+712	|	135625	|	family	|		|	0	|	1	|	11	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+724	|	712	|	genus	|		|	0	|	1	|	11	|	1	|	0	|	1	|	0	|	0	|	code compliant	|
+727	|	724	|	species	|	HI	|	0	|	1	|	11	|	1	|	0	|	1	|	1	|	0	|	code compliant; specified	|

modules/ebi-metagenomics/taxonkit/reformat/tests/main.nf.test

@@ -0,0 +1,61 @@
+nextflow_process {
+
+    name "Test Process TAXONKIT_REFORMAT"
+    script "../main.nf"
+    process "TAXONKIT_REFORMAT"
+
+    tag "modules"
+    tag "modules_ebimetagenomics"
+    tag "taxonkit"
+    tag "taxonkit/reformat"
+
+    config "./nextflow.config"
+
+    test("tsv with header") {
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file("$moduleTestDir/fixtures/nucl_gb.accession2taxid", checkIfExists: true)
+                ]
+                input[1] = file("$moduleTestDir/fixtures/taxdump", checkIfExists: true)
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+    test("tsv with header - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file("$moduleTestDir/fixtures/nucl_gb.accession2taxid", checkIfExists: true)
+                ]
+                input[1] = file("$moduleTestDir/fixtures/taxdump", checkIfExists: true)
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+            )
+        }
+
+    }
+
+}

modules/ebi-metagenomics/taxonkit/reformat/tests/main.nf.test.snap

@@ -0,0 +1,72 @@
+{
+    "tsv with header": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test.tsv:md5,331e1290ca8ae7e3962b43e28d865bae"
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,25e183506a3cbe3ddc98e6036b4a0525"
+                ],
+                "reformat_tsv": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test.tsv:md5,331e1290ca8ae7e3962b43e28d865bae"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,25e183506a3cbe3ddc98e6036b4a0525"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.0",
+            "nextflow": "24.04.3"
+        },
+        "timestamp": "2024-10-08T16:51:54.404921"
+    },
+    "tsv with header - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test.tsv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,25e183506a3cbe3ddc98e6036b4a0525"
+                ],
+                "reformat_tsv": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        "test.tsv:md5,d41d8cd98f00b204e9800998ecf8427e"
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,25e183506a3cbe3ddc98e6036b4a0525"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.0",
+            "nextflow": "24.04.3"
+        },
+        "timestamp": "2024-10-08T16:52:03.328778"
+    }
+}

modules/ebi-metagenomics/taxonkit/reformat/tests/nextflow.config

@@ -0,0 +1,3 @@
+process {
+    ext.args = '-t -r Unassigned -R -1 --taxid-field 3'
+}

modules/ebi-metagenomics/taxonkit/reformat/tests/tags.yml

@@ -0,0 +1,2 @@
+taxonkit/reformat:
+  - "modules/ebi-metagenomics/taxonkit/reformat/**"