annotate reldist.xml @ 35:b28e0cfa7ba1 draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/bedtools commit fe3f54a0d3edb83fcf6752e3b1524c582b4febd5"

<tool id="bedtools_reldistbed" name="bedtools ReldistBed" version="@TOOL_VERSION@">
    <description>calculate the distribution of relative distances</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <expand macro="stdio" />
    <command><![CDATA[
bedtools reldist
-a '$inputA'
-b '$inputB'
$detail
> '$output'
    ]]></command>
    <inputs>
        <param name="inputA" argument="-a" type="data" format="bam,@STD_BEDTOOLS_INPUTS@" label="@STD_BEDTOOLS_INPUT_LABEL@/BAM file"/>
        <param name="inputB" argument="-b" type="data" format="@STD_BEDTOOLS_INPUTS@" label="@STD_BEDTOOLS_INPUT_LABEL@ file"/>
        <param argument="-detail" type="boolean" truevalue="-detail" falsevalue="" checked="false"
            label="Instead of a summary, report the relative distance for each interval in A" />
    </inputs>
    <outputs>
        <data name="output" format_source="inputA" metadata_source="inputA" label="Relalative distance of ${inputA.name} and ${inputB.name}"/>
    </outputs>
    <tests>
        <test>
            <param name="inputA" value="a.bed" ftype="bed" />
            <param name="inputB" value="a.bed" ftype="bed" />
            <output name="output" file="reldistBed_result1.bed" ftype="bed" />
        </test>
    </tests>
    <help><![CDATA[
**What it does**

Traditional approaches to summarizing the similarity between two sets of genomic intervals are based upon the number or proportion of intersecting intervals. However, such measures are largely blind to spatial correlations between the two sets where, dpesite consistent spacing or proximity, intersections are rare (for example, enhancers and transcription start sites rarely overlap, yet they are much closer to one another than two sets of random intervals). Favorov et al proposed a relative distance metric that describes distribution of relative distances between each interval in one set nd the two closest intervals in another set (see figure above). If there is no spatial correlation between the two sets, one would expect the relative distances to be uniformaly distributed among the relative distances ranging from 0 to 0.5. If, however, the intervals tend to be much closer than expected by chance, the distribution of observed relative distances would be shifted towards low relative distance values (e.g., the figure below).

.. image:: $PATH_TO_IMAGES/reldist-glyph.png

.. class:: infomark

@REFERENCES@
    ]]></help>
    <expand macro="citations">
        <citation type="doi">10.1371/journal.pcbi.1002529</citation>
    </expand>
</tool>