name: gatk4_mutect2
description: Call somatic SNVs and indels via local assembly of haplotypes.
keywords:
  - gatk4
  - mutect2
  - haplotype
  - somatic
tools:
  - gatk4:
      description: |
        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/categories/360002369672s
      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']
  - input:
      type: list
      description: list of BAM files, also able to take CRAM as an input
      pattern: "*.{bam/cram}"
  - input_index:
      type: list
      description: list of BAM file indexes, also able to take CRAM indexes as an input
      pattern: "*.{bam.bai/cram.crai}"
  - intervals:
      type: File/string
      description: Specify region the tools is run on.
      pattern: ".{bed,interval_list}/chrM"
  - which_norm:
      type: list
      description: optional list of sample headers contained in the normal sample bam files (these are required for tumor_normal_pair mode)
      pattern: "testN"
  - run_single:
      type: boolean
      description: Specify whether or not to run in tumor_single mode instead of tumor_normal_pair mode (will be ignored if run_pon is also true)
      pattern: "true/false"
  - run_pon:
      type: boolean
      description: Specify whether or not to run in panel_of_normal mode instead of tumor_normal_pair mode
      pattern: "true/false"
  - run_mito:
      type: boolean
      description: Specify whether or not to run in mitochondria-mode instead of tumor_normal_pair mode
      pattern: "true/false"
  - fasta:
      type: file
      description: The reference fasta file
      pattern: "*.fasta"
  - fai:
      type: file
      description: Index of reference fasta file
      pattern: "*.fasta.fai"
  - dict:
      type: file
      description: GATK sequence dictionary
      pattern: "*.dict"
  - germline_resource:
      type: file
      description: Population vcf of germline sequencing, containing allele fractions.
      pattern: "*.vcf.gz"
  - germline_resource_tbi:
      type: file
      description: Index file for the germline resource.
      pattern: "*.vcf.gz.tbi"
  - panel_of_normals:
      type: file
      description: vcf file to be used as a panel of normals.
      pattern: "*.vcf.gz"
  - panel_of_normals_tbi:
      type: file
      description: Index for the panel of normals.
      pattern: "*.vcf.gz.tbi"

output:
  - vcf:
      type: file
      description: compressed vcf file
      pattern: "*.vcf.gz"
  - tbi:
      type: file
      description: Index of vcf file
      pattern: "*vcf.gz.tbi"
  - stats:
      type: file
      description: Stats file that pairs with output vcf file
      pattern: "*vcf.gz.stats"
  - f1r2:
      type: file
      description: file containing information to be passed to LearnReadOrientationModel (only outputted when tumor_normal_pair mode is run)
      pattern: "*.f1r2.tar.gz"
  - versions:
      type: file
      description: File containing software versions
      pattern: "versions.yml"

authors:
  - "@GCJMackenzie"