Merge pull request #53 from genomic-medicine-sweden/classification_kaiju

Add kaiju/kaiju in taxprofiler

README.md

@@ -44,6 +44,7 @@ On release, automated continuous integration tests run the pipeline on a full-si
    - [Centrifuge](https://ccb.jhu.edu/software/centrifuge/)
    - [Kaiju](https://kaiju.binf.ku.dk/)
    - [mOTUs](https://motu-tool.org/)
+   - [MetaMaps](https://github.com/DiltheyLab/MetaMaps)
 4. Perform optional post-processing with:
    - [bracken](https://ccb.jhu.edu/software/bracken/)
 5. Standardises output tables

conf/modules.config

@@ -237,7 +237,7 @@ process {
             pattern: '*.txt'
         ]
         ext.args = { "${meta.db_params}" }
-        ext.prefix = { "${meta.id}-${meta.run_accession}-${meta.db_name}" }
+        ext.prefix = params.perform_runmerging ? { "${meta.id}-${meta.db_name}" } : { "${meta.id}-${meta.run_accession}-${meta.db_name}" }
     }
 
     withName: CUSTOM_DUMPSOFTWAREVERSIONS {
@@ -256,4 +256,13 @@ process {
         ]
     }
 
+    withName: KAIJU_KAIJU {
+        publishDir = [
+            path: { "${params.outdir}/kaiju/${meta.db_name}" },
+            mode: params.publish_dir_mode,
+            pattern: '*.tsv'
+        ]
+        ext.args = { "${meta.db_params}" }
+        ext.prefix = params.perform_runmerging ? { "${meta.id}-${meta.db_name}" } : { "${meta.id}-${meta.run_accession}-${meta.db_name}" }
+    }
 }

conf/test.config

@@ -22,15 +22,16 @@ params {
     // Input data
     // TODO nf-core: Specify the paths to your test data on nf-core/test-datasets
     // TODO nf-core: Give any required params for the test so that command line flags are not needed
-    input                         = 'https://raw.githubusercontent.com/nf-core/test-datasets/taxprofiler/samplesheet.csv'
-    databases                     = 'https://raw.githubusercontent.com/nf-core/test-datasets/taxprofiler/database.csv'
-    run_kraken2                   = true
-    run_malt                      = true
-    run_metaphlan3                = true
-    run_centrifuge                = true
+    input                                 = 'https://raw.githubusercontent.com/nf-core/test-datasets/taxprofiler/samplesheet.csv'
+    databases                             = 'https://raw.githubusercontent.com/nf-core/test-datasets/taxprofiler/database.csv'
     perform_shortread_clipmerge           = true
     perform_longread_clip                 = false
     perform_shortread_complexityfilter    = true
     perform_shortread_hostremoval         = true
     shortread_hostremoval_reference       = 'https://raw.githubusercontent.com/nf-core/test-datasets/modules/data/genomics/homo_sapiens/genome/genome.fasta'
+    run_kaiju                             = true
+    run_kraken2                           = true
+    run_malt                              = true
+    run_metaphlan3                        = true
+    run_centrifuge                        = true
 }

docs/usage.md

@@ -124,6 +124,10 @@ Expected (uncompressed) database files for each tool are as follows:
   - `mpa_v30_CHOCOPhlAn_201901.rev.1.bt2`
   - `mpa_v30_CHOCOPhlAn_201901.rev.2.bt2`
   - `mpa_latest`
+- **Kaiju** output of `kaiju-makedb`. A directory containing:
+  - `kaiju_db_*.fmi`
+  - `nodes.dmp`
+  - `names.dmp`
 
 ## Running the pipeline
 

modules.json

@@ -53,6 +53,9 @@
             },
             "untar": {
                 "git_sha": "e080f4c8acf5760039ed12ec1f206170f3f9a918"
+            },
+            "kaiju/kaiju": {
+                "git_sha": "8856f127c58f6af479128be8b8df4d42e442ddbe"
             }
         }
     }

modules/nf-core/modules/kaiju/kaiju/main.nf

@@ -0,0 +1,41 @@
+process KAIJU_KAIJU {
+    tag "$meta.id"
+    label 'process_high'
+
+    conda (params.enable_conda ? "bioconda::kaiju=1.8.2" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/kaiju:1.8.2--h5b5514e_1':
+        'quay.io/biocontainers/kaiju:1.8.2--h5b5514e_1' }"
+
+    input:
+    tuple val(meta), path(reads)
+    path(db)
+
+    output:
+    tuple val(meta), path('*.tsv'), emit: results
+    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 input = meta.single_end ? "-i ${reads}" : "-i ${reads[0]} -j ${reads[1]}"
+    """
+    dbnodes=`find -L ${db} -name "*nodes.dmp"`
+    dbname=`find -L ${db} -name "*.fmi" -not -name "._*"`
+    kaiju \\
+        $args \\
+        -z $task.cpus \\
+        -t \$dbnodes \\
+        -f \$dbname \\
+        -o ${prefix}.tsv \\
+        $input
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        kaiju: \$(echo \$( kaiju -h 2>&1 | sed -n 1p | sed 's/^.*Kaiju //' ))
+    END_VERSIONS
+    """
+}

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

@@ -0,0 +1,53 @@
+name: kaiju_kaiju
+description: Taxonomic classification of metagenomic sequence data using a protein reference database
+keywords:
+  - classify
+  - metagenomics
+  - fastq
+  - taxonomic profiling
+tools:
+  - kaiju:
+      description: Fast and sensitive taxonomic classification for metagenomics
+      homepage: https://kaiju.binf.ku.dk/
+      documentation: https://github.com/bioinformatics-centre/kaiju/blob/master/README.md
+      tool_dev_url: https://github.com/bioinformatics-centre/kaiju
+      doi: "10.1038/ncomms11257"
+      licence: ["GNU 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/fasta files of size 1 and 2 for single-end and paired-end data,
+        respectively.
+      pattern: "*.{fastq,fq,fasta,fa,fsa,fas,fna,fastq.gz,fq.gz,fasta.gz,fa.gz,fsa.gz,fas.gz,fna.gz}"
+  - db:
+      type: files
+      description: |
+        List containing the database and nodes files for Kaiju
+        e.g. [ 'database.fmi', 'nodes.dmp' ]
+
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+  - results:
+      type: file
+      description: Results with taxonomic classification of each read
+      pattern: "*.tsv"
+
+authors:
+  - "@talnor"
+  - "@sofstam"
+  - "@jfy133"

nextflow.config

@@ -101,6 +101,9 @@ params {
 
     // metaphlan3
     run_metaphlan3             = false
+
+    // kaiju
+    run_kaiju                  = false
 }
 
 // Load base.config by default for all pipelines

nextflow_schema.json

@@ -370,6 +370,9 @@
             "type": "string",
             "default": "None"
         },
+        "run_kaiju": {
+            "type": "boolean"
+        },
         "malt_generatemegansummary": {
             "type": "boolean"
         }

subworkflows/local/profiling.nf

@@ -7,6 +7,7 @@ include { MEGAN_RMA2INFO              } from '../../modules/nf-core/modules/mega
 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'
+include { KAIJU_KAIJU                 } from '../../modules/nf-core/modules/kaiju/kaiju/main'
 
 workflow PROFILING {
     take:
@@ -37,6 +38,7 @@ workflow PROFILING {
                 kraken2: it[2]['tool'] == 'kraken2'
                 metaphlan3: it[2]['tool'] == 'metaphlan3'
                 centrifuge: it[2]['tool'] == 'centrifuge'
+                kaiju: it[2]['tool'] == 'kaiju'
                 unknown: true
             }
 
@@ -98,6 +100,13 @@ workflow PROFILING {
                                     db: it[3]
                             }
 
+    ch_input_for_kaiju = ch_input_for_profiling.kaiju
+                            .multiMap {
+                                it ->
+                                    reads: [it[0] + it[2], it[1]]
+                                    db: it[3]
+                            }
+
     /*
         RUN PROFILING
     */
@@ -143,6 +152,10 @@ workflow PROFILING {
         ch_raw_profiles    = ch_raw_profiles.mix( METAPHLAN3.out.biom )
     }
 
+    if ( params.run_kaiju ) {
+        KAIJU_KAIJU ( ch_input_for_kaiju.reads, ch_input_for_kaiju.db )
+        ch_versions = ch_versions.mix( KAIJU_KAIJU.out.versions.first() )
+    }
 
     emit:
     profiles = ch_raw_profiles    // channel: [ val(meta), [ reads ] ] - should be text files or biom