view overlapselect_wrapper.xml @ 0:84f5e02c76f3 draft default tip

Uploaded

<tool id="overlapselect_wrapper" name="overlap Select" version="1.0">
     <requirements>
       <requirement type='package' version="latest">overlapSelect</requirement>
     </requirements>


  <description>UCSC Overlap Select</description>
  <command interpreter="bash">
	overlapselect_wrapper.sh
#if $optionSpec.parType == "specify":
    #if $str($optionSpec.strand) == "true":
        -strand
    #end if
    #if $str($optionSpec.oppositeStrand) == "true":
        -oppositeStrand
    #end if
    #if $str($optionSpec.nonOverlapping) == "true":
        -nonOverlapping
    #end if
    #if $str($optionSpec.mergeOutput) == "true":
        -mergeOutput
    #end if
    -overlapBases=$optionSpec.overlapBases
    -overlapThreshold=$optionSpec.overlapThreshold
    -overlapThresholdCeil=$optionSpec.overlapThresholdCeil
    #if $str($optionSpec.selectFormat) == "bed":
        -selectFmt=bed
    #end if
    #if $str($optionSpec.inFormat) == "bed":
        -inFmt=bed
    #end if
    #if $str($optionSpec.selectFormat) == "tab":
        -selectCoordCols=$optionSpec.selColStart
    #end if
    #if $str($optionSpec.inFormat) == "tab":
        -inCoordCols=$optionSpec.inColStart
    #end if
#else
    -selectFmt=bed inFmt=bed
#end if
    $selectFile
    $inFile
    $outFile
  </command>
  <inputs>
    <param name="inFile" type="data" format="Tabular" metadata_name="dbkey" label="Select the query (input) file" />
    <param name="selectFile" type="data" format="Tabular" metadata_name="dbkey" label="Select the reference (select) file" />
    <conditional name="optionSpec">      
      <param name="parType" type="select" label="Do you want to select your target from the list or from your history?" help="">
        <option value="default">Use Default Parameters</option>
        <option value="specify">Specify Parameters</option>
      </param>
      <when value="default">
      </when>
      <when value="specify">
        <param name="selectFormat" type="select" label="Query File Format -selectFmt">
                          <option value="bed">Bed</option>
                          <option value="tab">Tabular</option>
        </param>
        <param name="inFormat" type="select" label="Reference File Format -inFmt">
                          <option value="bed">Bed</option>
                          <option value="tab">Tabular</option>
        </param>
        <param name="selColStart" type="integer" value="0" label="Query File is unstranded with chr starting at this column (0 based - i.e. 0=col1), followed by start and end (-1 to disable)-selectCoordCols" /> 
        <param name="inColStart" type="integer" value="0" label="Reference File is unstranded with chr starting at this column (0 based - i.e. 0=col1), followed by start and end  (-1 to disable)-inCoordCols" /> 
        <param name="strand" type="boolean" truevalue="true" falsevalue="false" checked="false" label="Only search on same strand -strand" />
        <param name="oppositeStrand" type="boolean" truevalue="true" falsevalue="false" checked="false" label="Only search on opposite strand -oppositeStrand" />
        <param name="nonOverlapping" type="boolean" truevalue="true" falsevalue="false" checked="false" label="Select non-overlapping instead of overlapping records" />
        <param name="mergeOutput" type="boolean" truevalue="true" falsevalue="false" checked="false" label="Merge Output -mergeOutput" />
        <param name="overlapBases" type="integer" value="-1" label="Number of overlap minimum (-overlapBases) (-1 to disable)" /> 
        <param name="overlapThreshold" type="float" value="0.0" label="Minimum fraction overlap" help="minimum fraction of an inFile record that must be overlapped by a single select record to be considered
        overlapping.  Note that this is only coverage by a single select record, not total coverage" /> 
        <param name="overlapThresholdCeil" type="float" value="1.1" label="Maximum fraction overlap" help="select only inFile records with less than this amount of overlap with a single record, provided they are selected by other criteria." /> 

      </when>  <!-- history -->	  	  
    </conditional>  <!-- optionSpec -->
  </inputs>
  <outputs>
    <data format="interval" name="outFile" label="Overlapselect on ${selectFile.name} vs ${inFile.name} "/>
  </outputs>
  <tests>
  	<test>
		<param name="inFile" value="NM_001206.gp"/>
		<param name="selectFile" value="mrna.psl"/>
		<output name="outFile" value="psl_over_NM_001206gp.psl"/>
  	</test>
  </tests>
  
<help>
**Overlap Select Help**


**Options**

* selectCds - Use only CDS in the selectFile

* selectRange - Use entire range instead of blocks from records in the selectFile.

* inFmt=fmt - specify inFile format, same values as -selectFmt.

* inCoordCols=spec - inFile is tab-separate with coordinates specified by spec, in format described above.

* inCds - Use only CDS in the inFile

* inRange - Use entire range instead of blocks of records in the inFile.

* nonOverlapping - select non-overlapping instead of overlapping records

* strand - must be on the same strand to be considered overlapping

* oppositeStrand - must be on the opposite strand to be considered overlapping

* excludeSelf - don't compare records with the same coordinates and name. Warning: using only one of -inCds or -selectCds will result in different coordinates for the same record.

* idMatch - only select overlapping records if they have the same id

* aggregate - instead of computing overlap bases on individual select entries, compute it based on the total number of inFile bases overlap by selectFile records. -overlapSimilarity and -mergeOutput will not work with this option.

* overlapThreshold=0.0 - minimum fraction of an inFile record that must be overlapped by a single select record to be considered overlapping.  Note that this is only coverage by a single select record, not total coverage.

* overlapThresholdCeil=1.1 - select only inFile records with less than this amount of overlap with a single record, provided they are selected by other criteria.

* overlapSimilarity=0.0 - minimum fraction of inFile and select records that: Note that this is only coverage by a single select record and this  is; bidirectional inFile and selectFile must overlap by this amount.  A value of 1.0 will select identical records (or CDS if  both CDS options are specified.  Not currently supported with  *aggregate.*

* overlapSimilarityCeil=1.1 - select only inFile records with less than this amount of similarity with a single record. provided they are selected by other criteria.

* overlapBases=-1 - minimum number of bases of overlap, lt 0 disables.

* statsOutput - output overlap statistics instead of selected records. If no overlap criteria is specified, all overlapping entries are reported, Otherwise only the pairs passing the criteria are reported. This results in a tab-separated file with the columns:  inId selectId inOverlap selectOverlap overBases: Where inOverlap is the fraction of the inFile record overlapped by the selectFile record and selectOverlap is the fraction of the select record overlap by inFile records.  With -aggregate, output is: inId inOverlap inOverBases inBases 

* statsOutputAll - like -statsOutput, however output all inFile records, including those that are not overlapped.

* statsOutputBoth - like -statsOutput, however output all selectFile and inFile records, including those that are not overlapped.

* mergeOutput - output file with be a merge of the input file with the selectFile records that selected it.  The format is inRec selectRec. if multiple select records hit, inRec is repeated. This will increase
      the memory required. Not supported with -nonOverlapping or -aggregate.
      
* idOutput - output a tab-separated file of pairs of inId selectId with -aggregate, only a single column of inId is written

* dropped=file  - output rows that were dropped to this file.

* verbose=n - verbose gt 1 prints some details,


  </help>
</tool>