name: "gatk4_collectreadcounts"
description: Collects read counts at specified intervals. The count for each interval is calculated by counting the number of read starts that lie in the interval.
keywords:
  - bam
  - cram
  - CollectReadCounts
  - gatk
  - gatk4
tools:
  - gatk4:
      description:
        Genome Analysis Toolkit (GATK4). Developed in the Data Sciences Platform at the Broad Institute, the toolkit offers a wide variety of tools
        with a primary focus on variant discovery and genotyping. Its powerful processing engine
        and high-performance computing features make it capable of taking on projects of any size.
      homepage: https://gatk.broadinstitute.org/hc/en-us
      documentation: https://gatk.broadinstitute.org/hc/en-us/articles/360037593911-CombineGVCFs
      tool_dev_url: https://github.com/broadinstitute/gatk
      doi: 10.1158/1538-7445.AM2017-3590
      licence: ["Apache-2.0"]

input:
  - meta:
      type: map
      description: |
        Groovy Map containing sample information
        e.g. [ id:'test', single_end:false ]
  - bam:
      type: file
      description: BAM/CRAM/SAM file
      pattern: "*.{bam,cram,sam}"
  - bai:
      type: file
      description: BAM/CRAM/SAM index file
      pattern: "*.{bai,crai,sai}"
  - intervals:
      type: file
      description: A file containing the specified intervals
      pattern: "*.{bed,intervals}"
  - fasta:
      type: file
      description: Optional - Reference FASTA
      pattern: "*.{fasta,fa}"
  - fai:
      type: file
      description: Optional - Index of the reference FASTA file
      pattern: "*.fai"
  - dict:
      type: file
      description: Optional - Sequence dictionary of the reference FASTA file
      pattern: "*.dict"

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"
  - hdf5:
      type: file
      description: The read counts in hdf5 format
      pattern: "*.hdf5"
  - tsv:
      type: file
      description: The read counts in TSV format
      pattern: "*.tsv"

authors:
  - "@nvnieuwk"