nf-core_modules/modules/dastool/dastool/meta.yml

name: dastool_dastool
description: DAS Tool binning step.
keywords:
  - binning
  - das tool
  - table
  - de novo
  - bins
  - contigs
  - assembly
  - das_tool
tools:
  - dastool:
      description: |
        DAS Tool is an automated method that integrates the results
        of a flexible number of binning algorithms to calculate an optimized, non-redundant
        set of bins from a single assembly.

      homepage: https://github.com/cmks/DAS_Tool
      documentation: https://github.com/cmks/DAS_Tool
      tool_dev_url: https://github.com/cmks/DAS_Tool
      doi: "10.1038/s41564-018-0171-1"
      licence: ["BSD"]

input:
  - meta:
      type: map
      description: |
        Groovy Map containing sample information
        e.g. [ id:'test', single_end:false ]
  - contigs:
      type: file
      description: fasta file
      pattern: "*.{fa.gz,fas.gz,fasta.gz}"
  - bins:
      type: file
      description: "Scaffolds2bin tabular file generated with dastool/scaffolds2bin"
      pattern: "*.scaffolds2bin.tsv"
  - proteins:
      type: file
      description: Predicted proteins in prodigal fasta format (>scaffoldID_geneNo)
      pattern: "*.{fa.gz,fas.gz,fasta.gz}"
  - db_directory:
      type: file
      description: (optional) Directory of single copy gene database.
  - search_engine:
      type: val
      description: Engine used for single copy gene identification. USEARCH is not supported due to it being proprietary [blast/diamond]

output:
  - meta:
      type: map
      description: |
        Groovy Map containing sample information
        e.g. [ id:'test', single_end:false ]
  - version:
      type: file
      description: File containing software version
      pattern: "versions.yml"
  - log:
      type: file
      description: Log file of the run
      pattern: "*.log"
  - summary:
      type: file
      description: Summary of output bins including quality and completeness estimates
      pattern: "*summary.txt"
  - scaffolds2bin:
      type: file
      description: Scaffolds to bin file of output bins
      pattern: "*.scaffolds2bin.txt"
  - eval:
      type: file
      description: Quality and completeness estimates of input bin sets
      pattern: "*.eval"
  - pdfs:
      type: file
      description: Plots showing the amount of high quality bins and score distribution of bins per method
      pattern: "*.pdf"
  - fasta_proteins:
      type: file
      description: Output from prodigal if not already supplied
      pattern: "*.proteins.faa"
  - fasta_archaea_scg:
      type: file
      description: Results of archaeal single-copy-gene prediction
      pattern: "*.archaea.scg"
  - fasta_bacteria_scg:
      type: file
      description: Results of bacterial single-copy-gene prediction
      pattern: "*.bacteria.scg"
  - seqlength:
      type: file
      description: Summary of contig lengths
      pattern: "*.seqlength"

authors:
  - "@maxibor"
  - "@jfy133"