<tool id="gemini_@BINARY@" name="GEMINI @BINARY@" version="@VERSION@.0">
    <description>Identifying potential compound heterozygotes</description>
    <expand macro="requirements" />
    <expand macro="version_command" />
    <macros>
        <import>gemini_macros.xml</import>
        <token name="@BINARY@">comp_hets</token>
    </macros>
    <command>
<![CDATA[
        gemini @BINARY@

            #if $report.report_selector != 'all':
                --columns "${report.columns}"
            #end if

            #if $filter.filter_selector == 'yes':
                --filter "${filter.filter}"
            #end if
            $only_affected
            $ignore_phasing

            "${ infile }"
            > "${ outfile }"
]]>
    </command>
    <expand macro="stdio" />
    <inputs>
        <param name="infile" type="data" format="sqlite" label="GEMINI database" />
        <expand macro="add_header_column" />
        <expand macro="column_filter" />
        <param name="only_affected" type="boolean" truevalue="--only-affected" falsevalue="" checked="False" 
            label="Report solely those compund heterozygotes impacted a sample labeled as affected" help="(--only-affected)"/>
        <param name="ignore_phasing" type="boolean" truevalue="--ignore-phasing" falsevalue="" checked="False" 
            label="Ignore phasing when screening for compound hets" help="Candidates are inherently putative. (--ignore-phasing)"/>
        <expand macro="filter" />
    </inputs>
    <outputs>
        <data name="outfile" format="tabular" label="${tool.name} on ${on_string}" />
    </outputs>
    <tests>
        <test>
        </test>
    </tests>
    <help>
**What it does**

Many recessive disorders are caused by compound heterozygotes. Unlike canonical recessive sites where the same recessive allele is 
inherited from both parents at the _same_ site in the gene, compound heterozygotes occur when the individual’s phenotype is caused 
by two heterozygous recessive alleles at _different_ sites in a particular gene.

So basically, we are looking for two (typically loss-of-function (LoF)) heterozygous variants impacting the same gene at different loci. 
The complicating factor is that this is _recessive_ and as such, we must also require that the consequential alleles at each heterozygous 
site were inherited on different chromosomes (one from each parent). As such, in order to use this tool, we require that all variants are phased. 
Once this has been done, the comp_hets tool will provide a report of candidate compound heterozygotes for each sample/gene.


@CITATION@
    </help>
    <expand macro="citations"/>
</tool>