<?xml version="1.0"?>

<tool id="concordance_report_1" name="Compute intra-NGS concordance">
  <description>over targeted regions in a genome</description>
  <version_string>kappa_report -v</version_string>
  <command interpreter="perl">kappa_report -q $target_bed $coding_gtf $input_ngs_calls1 $input_bam1 $coverage_cutoff1 $input_ngs_calls2 $input_bam2 $coverage_cutoff2 $out_summary_table</command>
  <inputs>
    <param format="gtf" name="coding_gtf" type="data" label="Coding regions" help="This must be the same (GTF) file used to define the coding regions for the HGVS annotation of the NGS variant calls"/>
    <param format="bed" name="target_bed" type="data" label="Target sequencing regions" help="e.g. a BED file from the Shared Data 'Exome target regions' folder, corresponding to the capture kit used for sequencing the sample"/>
    <param format="achri_snp_table" name="input_ngs_calls1" type="data" label="HGVS annotated NGS variant calls for genotyping method #1"/>
    <param format="bam" name="input_bam1" type="data" label="Mapped reads file (BAM) that was used to generate the NGS variant calls for genotyping method #1"/>
    <param type="integer" name="coverage_cutoff1" value="20" min="1" max="40" label="Report method #1 calls only for positions with read depth equal to or greater than..."/>
    <param format="achri_snp_table" name="input_ngs_calls2" type="data" label="HGVS annotated NGS variant calls for genotyping method #2"/>
    <param format="bam" name="input_bam2" type="data" label="Mapped reads file (BAM) that was used to generate the NGS variant calls for genotyping method #2"/>
    <param type="integer" name="coverage_cutoff2" value="20" min="1" max="40" label="Report method #2 calls only for positions with read depth equal to or greater than..."/>
  </inputs>
  <outputs>
    <data name="out_summary_table" format="tabular" label="Concordance summary (shared, unique calls, etc.)"/>
  </outputs>

  <tests>
  </tests>

  <help>
**What it does**

This tool reports several statistics comparing the genotypes derived from two next-gen sequencing runs. This is intended to measure methodological consistency,
assuming the samples for both methods are from the same individual. When coverage is good (typically 10x for colorspace, 20x for base space), there should be a large
(>= 95%) overlap between the calls if they are reasonably accurate.
  </help>

</tool>