view hicQuickQC.xml @ 0:269c4246279d draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/hicexplorer commit 3b41d687ff30583540d055f6995de00530cca81d-dirty"

<tool id="hicexplorer_hicquickqc" name="@BINARY@" version="@WRAPPER_VERSION@.0">
    <description>get a first quality estimate of Hi-C data</description>
    <macros>
        <token name="@BINARY@">hicQuickQC</token>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
   
        mkdir ./QCfolder &&
        mkdir $qc.files_path &&
        @BINARY@

            --samFiles
            #for $repeat in $samFiles:
                '${repeat.samFile}'
            #end for

            #if $restrictionCutFileBinSize_conditional.restrictionCutFileBinSize_selector == "optionRestrictionCutFile":
                --restrictionCutFile '$restrictionCutFileBinSize_conditional.restrictionCutFile'
                --minDistance $restrictionCutFileBinSize_conditional.minDistance
                --maxLibraryInsertSize $restrictionCutFileBinSize_conditional.maxLibraryInsertSize
            #end if

            #if $restrictionCutFileBinSize_conditional.restrictionCutFileBinSize_selector == "optionBinSize":
                --binSize 
                #for $repeat in $restrictionCutFileBinSize_conditional.binSizes
                    '${repeat.binSize}'
                #end for
            #end if


            #if $restrictionSequence:
                --restrictionSequence '$restrictionSequence'
            #end if
            
            #if $danglingSequence:
                --danglingSequence '$danglingSequence'
            #end if
            --QCfolder ./QCfolder

            --lines $lines

        && mv ./QCfolder/* $qc.files_path/
        && mv $qc.files_path/hicQC.html $qc
        && mv $qc.files_path/*.log raw_qc
    ]]></command>
    <inputs>
         <repeat max="2" min="2" name="samFiles" title="Sam/Bam files to process (forward/reverse)" 
                help="Please use the special BAM datatype: qname_input_sorted.bam and use for 'bowtie2' the '--reorder' option to create a BAM file.">
            <param name="samFile" type="data" format="sam,qname_input_sorted.bam"/>
        </repeat>
        <conditional name="restrictionCutFileBinSize_conditional">
            <param name="restrictionCutFileBinSize_selector" type="select" label="Choose to use a restriction cut file or a bin size">
                <option value="optionRestrictionCutFile">Restriction cut file</option>
                <option value="optionBinSize" selected="True">Bin size</option>
            </param>
            <when value="optionRestrictionCutFile">
                <param argument="--restrictionCutFile" type="data" format="bed" optional="true" label="BED file with all restriction cut places"
                        help="Should contaion only  mappable restriction sites. If given, the bins are set to match the restriction fragments
                        (i.e. the region between one restriction site and the next)." />
                <param argument="--minDistance" type="integer" value="" optional="true" label="Minimum distance between restriction sites"
                        help="Restriction sites that are closer that this distance are merged into one.
                        This option only applies if --restrictionCutFile is given."/>
                <param argument="--maxLibraryInsertSize" type="integer" value="" optional="true"
                        label="Maximum library insert size defines different cut offs based on the maximum expected library size"
                        help="*This is not the average fragment size* but the higher end of the fragment size distribution (obtained using for example Fragment Analyzer)
                              which usually is between 800 to 1500 bp. If this value if not known use the default of 1000. The insert value is used to decide if two mates
                              belong to the same fragment (by checking if they are within this max insert size) and to decide if a mate
                              is too far away from the nearest restriction site." />
            </when>
            <when value="optionBinSize">
                <repeat  name='binSizes' title='Bin size in bp' min="1" help="If used, the restriction cut places (if given) are used to only consider reads that are in the vicinity of the resctriction sites.
                        Otherwise all reads in the interval are considered. Use multiple ones to create a mcool file.">
                    <param argument="--binSize" type="integer" value="" optional="true" label="Bin size in bp"/>
                </repeat>
            </when>
        </conditional>

        <param argument="--restrictionSequence" type="text" optional="true" label="Sequence of the restriction site"
            help="This is used to discard reads that end/start with such sequence and that are considered un-ligated fragments or
            &quot;dangling-ends&quot;. If not given, such statistics will not be available." />

        <param argument="--danglingSequence" type="text" optional="true" label="Dangling sequence"
            help="Sequence left by the restriction enzyme after cutting.
                    Each restriction enzyme recognizes a different DNA sequence and, 
                    after cutting, they leave behind a specific ‘sticky’ end or dangling end sequence.
                    For example, for HindIII the restriction site is AAGCTT and the dangling end is AGCT. 
                    For DpnII, the restriction site and dangling end sequence are the same: GATC. 
                    This information is easily found on the description of the restriction enzyme.
                    The dangling sequence is used to classify and report reads whose 5’ end starts with such sequence as dangling-end reads.
                    A significant portion of dangling-end reads in a sample are indicative of a problem with the re-ligation step of the protocol. "/>

        <param argument="--lines" optional='true'  type="integer" label="Lines" help= "Number of lines to consider for the qc test run." value='1000000'/>
        
    </inputs>
    <outputs>
       <data name="qc" format="html" label="${tool.name} QC on ${on_string}"/>
       <data name="raw_qc" from_work_dir='raw_qc' format='txt' label="${tool.name} raw QC on ${on_string}" />
    </outputs>
    <tests>
        <test>
            <repeat name="samFiles">
                <param name="samFile" value="small_test_R1_unsorted.sam"/>
            </repeat>
            <repeat name="samFiles">
                <param name="samFile" value="small_test_R2_unsorted.sam"/>
            </repeat>
            <conditional name="restrictionCutFileBinSize_conditional">
                <param name="restrictionCutFileBinSize_selector" value="optionBinSize"/>
                <repeat name='binSizes'>
                    <param name="binSize" value="5000"/>
                </repeat>
            </conditional>
            <param name="lines" value='1000'/>
            <output name="raw_qc" file='hicQuickQC/QC.log' compare='diff' lines_diff='2'/>
        </test>
    </tests>
    <help><![CDATA[

Quick QC 
====================

Get a quick impression on the quality of your Hi-C data. The first 

For more information about HiCExplorer please consider our documentation on readthedocs.io_

.. _readthedocs.io: http://hicexplorer.readthedocs.io/en/latest/index.html
]]></help>
    <expand macro="citations" />
</tool>