<tool id="gemini_@BINARY@" name="GEMINI @BINARY@" version="@VERSION@.1">
    <description>Identifying runs of homozygosity</description>
    <macros>
        <import>gemini_macros.xml</import>
        <token name="@BINARY@">roh</token>
    </macros>
    <expand macro="requirements" />
    <expand macro="stdio" />
    <expand macro="version_command" />
    <command>
<![CDATA[
        gemini @BINARY@
            --min-snps $min_snps
            --min-total-depth $min_total_depth
            --min-gt-depth $min_gt_depth
            --min-size $min_size
            --max-hets $max_hets
            --max-unknowns $max_unknowns
            #if $samples.strip():
                -s "${samples}"
            #end if
            "${ infile }"
            > "${ outfile }"
]]>
    </command>
    <inputs>
        <expand macro="infile" />

        <param name="min_snps" type="integer" value="25" label="Minimum number of expected homozygous SNPs" help="default: 25 (--min-snps)">
            <validator type="in_range" min="0"/>
        </param>
        <param name="min_total_depth" type="integer" value="20" label="The minimum overall sequencing depth requiredfor a SNP to be considered" help="default: 20 (--min-total-depth)">
            <validator type="in_range" min="0"/>
        </param>
        <param name="min_gt_depth" type="integer" value="0" label="The minimum required sequencing depth underlying a given sample's genotype for a SNP to be considered"
            help="default: 0 (--min-gt-depth)">
            <validator type="in_range" min="0"/>
        </param>
        <param name="min_size" type="integer" value="100000" label="Minimum run size in base pairs" help="default: 100000 (--min-size)">
            <validator type="in_range" min="1"/>
        </param>
        <param name="max_hets" type="integer" value="1" label="Maximum number of allowed hets in the run" help="default: 1 (--max-hets)">
            <validator type="in_range" min="1"/>
        </param>
        <param name="max_unknowns" type="integer" value="3" label="Maximum number of allowed unknowns in the run" help="default: 3 (-max-unknowns)">
            <validator type="in_range" min="0"/>
        </param>

        <param name="samples" type="text" value="" label="Comma separated list of samples to screen for ROHs" help="e.g S120,S450 (-s)"/>

    </inputs>

    <outputs>
        <data name="outfile" format="tabular" />
    </outputs>
    <tests>
        <test>
            <param name="infile" value="gemini_load_result1.db" ftype="gemini.sqlite" />
            <param name="min_snps" value="3" />
            <param name="min_size" value="10" />
            <param name="min_total_depth" value="0" />
            <output name="outfile">
                <assert_contents>
                    <has_line_matching expression="chrom&#009;start&#009;end&#009;.*run_length.*" />
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[

**What it does**

===========================================================================
``ROH``: Identifying runs of homozygosity
===========================================================================
Runs of homozygosity are long stretches of homozygous genotypes that reflect
segments shared identically by descent and are a result of consanguinity or
natural selection. Consanguinity elevates the occurrence of rare recessive
diseases (e.g. cystic fibrosis) that represent homozygotes for strongly deleterious
mutations. Hence, the identification of these runs holds medical value.

The 'roh' tool in GEMINI returns runs of homozygosity identified in whole genome data.
The tool basically looks at every homozygous position on the chromosome as a possible
start site for the run and looks for those that could give rise to a potentially long
stretch of homozygous genotypes.

For e.g. for the given example allowing ``1 HET`` genotype (h) and ``2 UKW`` genotypes (u)
the possible roh runs (H) would be:


::

	genotype_run = H H H H h H H H H u H H H H H u H H H H H H H h H H H H H h H H H H H
	roh_run1     = H H H H h H H H H u H H H H H u H H H H H H H
	roh_run2     =           H H H H u H H H H H u H H H H H H H h H H H H H
	roh_run3     =                     H H H H H u H H H H H H H h H H H H H
	roh_run4     =                                 H H H H H H H h H H H H H

roh returned for --min-snps = 20 would be:

::

	roh_run1     = H H H H h H H H H u H H H H H u H H H H H H H
	roh_run2     =           H H H H u H H H H H u H H H H H H H h H H H H H


As you can see, the immediate homozygous position right of a break (h or u) would be the possible
start of a new roh run and genotypes to the left of a break are pruned since they cannot
be part of a longer run than we have seen before.


    ]]></help>
    <expand macro="citations"/>
</tool>