<?xml version="1.0"?>

<tool id="microarray_reports_1" name="Compute NGS/microarray concordance">
  <description>over targeted regions in a genome</description>
  <version_string>microarray_reports -v</version_string>
  <command interpreter="perl">microarray_reports $__tool_data_path__ -q $out_discordant_bed $out_summary_table $input_ngs_calls $input_micro_calls $input_bam $target_gtf $ref_genome.fa $coverage_cutoff $target_bed</command>
  <inputs>
    <param format="achri_snp_table" name="input_ngs_calls" type="data" label="HGVS annotated NGS variant calls"/>
    <param format="tabular" name="input_micro_calls" type="data" label="Microarray genotype calls. See help info below this form for required input format."/>
    <param format="bam" name="input_bam" type="data" label="Mapped reads file (BAM) that was used to generate the NGS variant calls"/>
    <param type="integer" name="coverage_cutoff" value="9" min="0" max="40" label="Report only positions with read depth greater than..."/>
    <param name="ref_genome" type="genomebuild" label="Reference genome" help="against which the NGS variants were called"/>
    <param format="gtf" name="target_gtf" type="data" label="Target 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 the
 to the capture kit used for sequencing the sample"/>
  </inputs>
  <outputs>
    <data name="out_summary_table" format="tabular" label="Concordance summary (false positive rate, etc.)"/>
    <data name="out_discordant_bed" format="bed" label="Discordant call locations"/>
  </outputs>

  <tests>
  </tests>

  <help>
**What it does**

This tool reports several statistics comparing the genotypes derived from a next-gen sequencing run to a microarray genotyping result 
for the same individual. When coverage is good (10x for colorspace, 20x for base space), false positive and false negative rates should 
be below 2%, assuming 99% accuracy of both the microarray and the NGS genotyping.

------

** Example**

The microarray file should have five columns (ProbeID, forward strand genotypes, dbSNP ID, chr, pos), e.g.::

    #Comment lines...
    Probe set header line
    SNP_A-1780520	GG	rs16994928	20	48440771
    SNP_A-1780985	AG	rs3859360	18	34178190
  </help>

</tool>