Merge pull request #31 from genomic-medicine-sweden/classification_centrifuge

Add centrifuge classification (centrifuge/centrifuge module)

conf/modules.config

@@ -198,4 +198,14 @@ process {
         ]
     }
 
+    withName: CENTRIFUGE_CENTRIFUGE {
+        publishDir = [
+            path: { "${params.outdir}/centrifuge/${meta.db_name}" },
+            mode: params.publish_dir_mode,
+            pattern: '*.{fastq.gz,txt}'
+        ]
+        ext.args = { "${meta.db_params}" }
+        ext.prefix = { "${meta.id}-${meta.run_accession}-${meta.db_name}" }
+    }
+
 }

conf/test.config

@@ -27,6 +27,7 @@ params {
     run_kraken2                   = true
     run_malt                      = true
     run_metaphlan3                = true
+    run_centrifuge                = true
     shortread_clipmerge           = true
     longread_clip                 = false
     shortread_complexityfilter    = true

modules.json

@@ -12,6 +12,9 @@
             "cat/fastq": {
                 "git_sha": "e745e167c1020928ef20ea1397b6b4d230681b4d"
             },
+            "centrifuge/centrifuge": {
+                "git_sha": "d2726fcf75063960f06b36d2229a4c0966614108"
+            },
             "custom/dumpsoftwareversions": {
                 "git_sha": "e745e167c1020928ef20ea1397b6b4d230681b4d"
             },

modules/nf-core/modules/centrifuge/centrifuge/main.nf

@@ -0,0 +1,61 @@
+process CENTRIFUGE_CENTRIFUGE {
+    tag "$meta.id"
+    label 'process_high'
+
+    conda (params.enable_conda ? "bioconda::centrifuge=1.0.4_beta" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/centrifuge:1.0.4_beta--h9a82719_6' :
+        'quay.io/biocontainers/centrifuge:1.0.4_beta--h9a82719_6' }"
+
+    input:
+    tuple val(meta), path(reads)
+    path db
+    val save_unaligned
+    val save_aligned
+    val sam_format
+
+    output:
+    tuple val(meta), path('*report.txt')                 , emit: report
+    tuple val(meta), path('*results.txt')                , emit: results
+    tuple val(meta), path('*.sam')                       , optional: true, emit: sam
+    tuple val(meta), path('*.mapped.fastq{,.1,.2}.gz')   , optional: true, emit: fastq_mapped
+    tuple val(meta), path('*.unmapped.fastq{,.1,.2}.gz') , optional: true, emit: fastq_unmapped
+    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 paired = meta.single_end ? "-U ${reads}" :  "-1 ${reads[0]} -2 ${reads[1]}"
+    def unaligned = ''
+    def aligned = ''
+    if (meta.single_end) {
+        unaligned = save_unaligned ? "--un-gz ${prefix}.unmapped.fastq.gz" : ''
+        aligned = save_aligned ? "--al-gz ${prefix}.mapped.fastq.gz" : ''
+    } else {
+        unaligned = save_unaligned ? "--un-conc-gz ${prefix}.unmapped.fastq.gz" : ''
+        aligned = save_aligned ? "--al-conc-gz ${prefix}.mapped.fastq.gz" : ''
+    }
+    def sam_output = sam_format ? "--out-fmt 'sam'" : ''
+    """
+    ## we add "-no-name ._" to ensure silly Mac OSX metafiles files aren't included
+    db_name=`find -L ${db} -name "*.1.cf" -not -name "._*"  | sed 's/.1.cf//'`
+    centrifuge \\
+        -x \$db_name \\
+        -p $task.cpus \\
+        $paired \\
+        --report-file ${prefix}.report.txt \\
+        -S ${prefix}.results.txt \\
+        $unaligned \\
+        $aligned \\
+        $sam_output \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        centrifuge: \$( centrifuge --version  | sed -n 1p | sed 's/^.*centrifuge-class version //')
+    END_VERSIONS
+    """
+}

modules/nf-core/modules/centrifuge/centrifuge/meta.yml

@@ -0,0 +1,66 @@
+name: centrifuge_centrifuge
+description: Classifies metagenomic sequence data
+keywords:
+  - classify
+  - metagenomics
+  - fastq
+  - db
+tools:
+  - centrifuge:
+      description: Centrifuge is a classifier for metagenomic sequences.
+      homepage: https://ccb.jhu.edu/software/centrifuge/
+      documentation: https://ccb.jhu.edu/software/centrifuge/manual.shtml
+      doi: 10.1101/gr.210641.116
+      licence: ["GPL v3"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - reads:
+      type: file
+      description: |
+        List of input FastQ files of size 1 and 2 for single-end and paired-end data,
+        respectively.
+  - db:
+      type: directory
+      description: Path to directory containing centrifuge database files
+  - save_unaligned:
+      type: value
+      description: If true unmapped fastq files are saved
+  - save_aligned:
+      type: value
+      description: If true mapped fastq files are saved
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - report:
+      type: file
+      description: |
+        File containing a classification summary
+      pattern: "*.{report.txt}"
+  - results:
+      type: file
+      description: |
+        File containing classification results
+      pattern: "*.{results.txt}"
+  - fastq_unmapped:
+      type: file
+      description: Unmapped fastq files
+      pattern: "*.unmapped.fastq.gz"
+  - fastq_mapped:
+      type: file
+      description: Mapped fastq files
+      pattern: "*.mapped.fastq.gz"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@sofstam"
+  - "@jfy133"
+  - "@sateeshperi"

nextflow.config

@@ -84,6 +84,11 @@ params {
     // kraken2
     run_kraken2                = false
 
+    // centrifuge
+    run_centrifuge             = false
+    centrifuge_save_unaligned  = false
+    centrifuge_save_aligned    = false
+    centrifuge_sam_format      = false
     // metaphlan3
     run_metaphlan3             = false
 }

nextflow_schema.json

@@ -281,6 +281,18 @@
         "run_kraken2": {
             "type": "boolean"
         },
+        "run_centrifuge": {
+            "type": "boolean"
+        },
+        "centrifuge_save_unaligned": {
+            "type": "boolean"
+        },
+        "centrifuge_save_aligned": {
+            "type": "boolean"
+        },
+        "centrifuge_sam_format": {
+            "type": "boolean"
+        },
         "run_metaphlan3": {
             "type": "boolean",
             "description": "Enable MetaPhlAn for taxonomic profiling"

nf-core/modules/centrifuge/centrifuge/main.nf

@@ -0,0 +1,61 @@
+process CENTRIFUGE_CENTRIFUGE {
+    tag "$meta.id"
+    label 'process_high'
+
+    conda (params.enable_conda ? "bioconda::centrifuge=1.0.4_beta" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/centrifuge:1.0.4_beta--h9a82719_6' :
+        'quay.io/biocontainers/centrifuge:1.0.4_beta--h9a82719_6' }"
+
+    input:
+    tuple val(meta), path(reads)
+    path db
+    val save_unaligned
+    val save_aligned
+    val sam_format
+
+    output:
+    tuple val(meta), path('*report.txt')                 , emit: report
+    tuple val(meta), path('*results.txt')                , emit: results
+    tuple val(meta), path('*.sam')                       , optional: true, emit: sam
+    tuple val(meta), path('*.mapped.fastq{,.1,.2}.gz')   , optional: true, emit: fastq_mapped
+    tuple val(meta), path('*.unmapped.fastq{,.1,.2}.gz') , optional: true, emit: fastq_unmapped
+    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 paired = meta.single_end ? "-U ${reads}" :  "-1 ${reads[0]} -2 ${reads[1]}"
+    def unaligned = ''
+    def aligned = ''
+    if (meta.single_end) {
+        unaligned = save_unaligned ? "--un-gz ${prefix}.unmapped.fastq.gz" : ''
+        aligned = save_aligned ? "--al-gz ${prefix}.mapped.fastq.gz" : ''
+    } else {
+        unaligned = save_unaligned ? "--un-conc-gz ${prefix}.unmapped.fastq.gz" : ''
+        aligned = save_aligned ? "--al-conc-gz ${prefix}.mapped.fastq.gz" : ''
+    }
+    def sam_output = sam_format ? "--out-fmt 'sam'" : ''
+    """
+    ## we add "-no-name ._" to ensure silly Mac OSX metafiles files aren't included
+    db_name=`find -L ${db} -name "*.1.cf" -not -name "._*"  | sed 's/.1.cf//'`
+    centrifuge \\
+        -x \$db_name \\
+        -p $task.cpus \\
+        $paired \\
+        --report-file ${prefix}.report.txt \\
+        -S ${prefix}.results.txt \\
+        $unaligned \\
+        $aligned \\
+        $sam_output \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        centrifuge: \$( centrifuge --version  | sed -n 1p | sed 's/^.*centrifuge-class version //')
+    END_VERSIONS
+    """
+}

nf-core/modules/centrifuge/centrifuge/meta.yml

@@ -0,0 +1,66 @@
+name: centrifuge_centrifuge
+description: Classifies metagenomic sequence data
+keywords:
+  - classify
+  - metagenomics
+  - fastq
+  - db
+tools:
+  - centrifuge:
+      description: Centrifuge is a classifier for metagenomic sequences.
+      homepage: https://ccb.jhu.edu/software/centrifuge/
+      documentation: https://ccb.jhu.edu/software/centrifuge/manual.shtml
+      doi: 10.1101/gr.210641.116
+      licence: ["GPL v3"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - reads:
+      type: file
+      description: |
+        List of input FastQ files of size 1 and 2 for single-end and paired-end data,
+        respectively.
+  - db:
+      type: directory
+      description: Path to directory containing centrifuge database files
+  - save_unaligned:
+      type: value
+      description: If true unmapped fastq files are saved
+  - save_aligned:
+      type: value
+      description: If true mapped fastq files are saved
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - report:
+      type: file
+      description: |
+        File containing a classification summary
+      pattern: "*.{report.txt}"
+  - results:
+      type: file
+      description: |
+        File containing classification results
+      pattern: "*.{results.txt}"
+  - fastq_unmapped:
+      type: file
+      description: Unmapped fastq files
+      pattern: "*.unmapped.fastq.gz"
+  - fastq_mapped:
+      type: file
+      description: Mapped fastq files
+      pattern: "*.mapped.fastq.gz"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@sofstam"
+  - "@jfy133"
+  - "@sateeshperi"

subworkflows/local/db_check.nf

@@ -22,7 +22,7 @@ workflow DB_CHECK {
 
     ch_dbs_for_untar = parsed_samplesheet
         .branch {
-            untar: it[1].toString().endsWith(".tar.gz") && it[0]['tool'] != 'centrifuge'
+            untar: it[1].toString().endsWith(".tar.gz")
             skip: true
         }
 

subworkflows/local/input_check.nf

@@ -63,13 +63,12 @@ def create_fastq_channel(LinkedHashMap row) {
             if (!file(row.fastq_2).exists()) {
                 exit 1, "ERROR: Please check input samplesheet -> Read 2 FastQ file does not exist!\n${row.fastq_2}"
             }
-            fastq_meta = [ meta, [ file(row.fastq_1), file(row.fastq_2) ] ]
+         fastq_meta = [ meta, [ file(row.fastq_1), file(row.fastq_2) ] ]
         }
+
     }
     return fastq_meta
-}
-
-// Function to get list of [ meta, fasta ]
+}// Function to get list of [ meta, fasta ]
 def create_fasta_channel(LinkedHashMap row) {
     def meta = [:]
     meta.id                     = row.sample

workflows/taxprofiler.nf

@@ -61,6 +61,7 @@ include { CUSTOM_DUMPSOFTWAREVERSIONS } from '../modules/nf-core/modules/custom/
 include { CAT_FASTQ                   } from '../modules/nf-core/modules/cat/fastq/main'
 include { MALT_RUN                    } from '../modules/nf-core/modules/malt/run/main'
 include { KRAKEN2_KRAKEN2             } from '../modules/nf-core/modules/kraken2/kraken2/main'
+include { CENTRIFUGE_CENTRIFUGE       } from '../modules/nf-core/modules/centrifuge/centrifuge/main'
 include { METAPHLAN3                  } from '../modules/nf-core/modules/metaphlan3/main'
 
 /*
@@ -137,6 +138,7 @@ workflow TAXPROFILER {
                 malt:    it[2]['tool'] == 'malt'
                 kraken2: it[2]['tool'] == 'kraken2'
                 metaphlan3: it[2]['tool'] == 'metaphlan3'
+                centrifuge: it[2]['tool'] == 'centrifuge'
                 unknown: true
             }
 
@@ -170,6 +172,18 @@ workflow TAXPROFILER {
                                     db: it[3]
                             }
 
+    // We can run centrifuge one-by-one sample-wise
+    ch_input_for_centrifuge =  ch_input_for_profiling.centrifuge
+                               .dump(tag: "input for centrifuge")
+                               .multiMap {
+                                    it ->
+                                        reads: [ it[0] + it[2], it[1] ]
+                                        db: it[3]
+                                }
+
+    //
+    // RUN PROFILING
+    //
     ch_input_for_metaphlan3 = ch_input_for_profiling.metaphlan3
                             .multiMap {
                                 it ->
@@ -188,6 +202,10 @@ workflow TAXPROFILER {
         KRAKEN2_KRAKEN2 ( ch_input_for_kraken2.reads, ch_input_for_kraken2.db  )
     }
 
+    if ( params.run_centrifuge ) {
+        CENTRIFUGE_CENTRIFUGE ( ch_input_for_centrifuge.reads, ch_input_for_centrifuge.db, params.centrifuge_save_unaligned, params.centrifuge_save_aligned, params.centrifuge_sam_format  )
+    }
+
     if ( params.run_metaphlan3 ) {
         METAPHLAN3 ( ch_input_for_metaphlan3.reads, ch_input_for_metaphlan3.db )
     }