<tool id="bedtools_groupbybed" name="bedtools GroupByBed" version="@TOOL_VERSION@" profile="@PROFILE@">
    <description>group by common cols and summarize other cols</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="bio_tools" />
    <expand macro="requirements" />
    <expand macro="stdio" />
    <command><![CDATA[
bedtools groupby
-i '${inputA}'
-g '$group'
-c '${cols}'
-o $operation
> '${output}'
    ]]></command>
    <inputs>
        <param name="inputA" argument="-i" type="data" format="@STD_BEDTOOLS_INPUTS@" label="@STD_BEDTOOLS_INPUT_LABEL@ file"/>
        <expand macro="choose_columns" />
        <param name="group" argument="-g" type="text" value="1,2,3"
            label="Specifies which column(s) (1-based) should be used to group the input"
            help="Columns may be comma-separated with each column must be explicitly listed. Or, ranges (e.g. 1-4) are also allowed.">
            <sanitizer invalid_char="">
                <valid initial="string.digits"><add value=","/><add value="-"/></valid>
            </sanitizer>
        </param>
        <param name="operation" argument="-o" type="select" label="Specify the operation">
            <!-- <option value="sum" selected="true">Sum - numeric only</option> This is already in math-->
            <option value="stdev">Stdev - numeric only</option>
            <option value="sstdev">Sstdev - numeric only</option>
            <option value="freqasc">Freqasc - comma separated list of values observed and the number of times they were observed (ascending)</option>
            <option value="freqdesc">Freqdesc - comma separated list of values observed and the number of times they were observed (descending)</option>
            <option value="first">First - numeric or text</option>
            <option value="last">Last - numeric or text</option>
            <expand macro="math_options" />
            <expand macro="additional_math_options" />
        </param>
    </inputs>
    <outputs>
        <data name="output" format_source="inputA" metadata_source="inputA"/>
    </outputs>
    <tests>
        <test>
            <param name="inputA" value="groupbyBed1.bed" ftype="bed" />
            <param name="cols" value="9" />
            <param name="group" value="1,2,3" />
            <param name="operation" value="sum" />
            <output name="output" file="groupbyBed_result1.bed" ftype="bed" />
        </test>
        <test>
            <param name="inputA" value="groupbyBed1.bed" ftype="bed" />
            <param name="cols" value="9" />
            <param name="group" value="1,2,3" />
            <param name="operation" value="min" />
            <output name="output" file="groupbyBed_result2.bed" ftype="bed" />
        </test>
        <test>
            <param name="inputA" value="groupbyBed1.bed" ftype="bed" />
            <param name="cols" value="9" />
            <param name="group" value="1-4" />
            <param name="operation" value="median" />
            <output name="output" file="groupbyBed_result3.bed" ftype="bed" />
        </test>
    </tests>
    <help><![CDATA[
**What it does**

Replicate lines in a file based on columns of comma-separated values.

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