view delly.xml @ 27:5d9e545c6ffb draft

Uploaded
author malbuquerque
date Tue, 20 Jan 2015 18:53:15 -0500
parents 226f241f0c92
children
line wrap: on
line source

<tool id="delly" name="DELLY" version="0.6.1">

	<description>structural variant discovery by integrated paired-end and split-read analysis</description>

	<requirements>
		<requirement type="package" version="0.6.1">delly</requirement>
		<requirement type="set_environment">DELLY_DIR</requirement>
	</requirements>


	<command> 

		<!-- BAM and BAI linking, (1) link BAM to new BAM file & (2) link BAM metadata to new BAI file -->
		#for $i, $s in enumerate( $repeatBam )
			ln -s $s.sortedBam ./input$(i).bam;
			ln -s $s.sortedBam.metadata.bam_index ./input$(i).bam.bai;
		#end for		

		<!-- Sets args to a list of types selected -->
		#if not isinstance( $type.value, list ):
			#set $args = [ $type.value ]
		#else:
			#set $args = $type.value
		#end if

		<!-- Run Delly Jobs for each type selected -->
		#for $option in $args
			<!-- NEED TO FIX -->
			<!-- Delly should be automatically installed into the galaxy instance and should not be running off 
					 the computers specific tool set -->
			\$DELLY_DIR/src/delly -t $option -o ./output.$(option).vcf -q $mapQual -s $madCutoff 
			#if $option == "DEL" 
			-m $minFlank 
			#end if 
			-u $genoQual -v $vcfgeno -g $genome
			
			<!-- add each input bam to command -->
			#for $i, $s in enumerate( $repeatBam )
				./input$(i).bam	
			#end for
			;
  	#end for

		<!-- Combine VCF Files and Sort Lexographically -->
		#set $option = $args[0]
		grep ^\# output.$(option).vcf > $outfile;
		grep ^\# -v output.$(option).vcf > variants.txt;
		
		<!-- If we called more than a single variant type, concatenate all the other types variant output -->
		#if isinstance( $type.value, list ):
			#for $option in $args[1:]
				grep ^\# -v output.$(option).vcf >> variants.txt;
			#end for
		#end if
		
		<!-- Sort all variant output, assuming that it will sort lexographically by chromosome, then position, ID  -->
		<!-- In future, maybe develop a script to sort by bam header -->
		sort -k1,1d -k2,2n -k3,3d variants.txt > sortedVariants.txt;
		
		<!-- Filter Variants that have Passed Quality Checks -->
		#if $filterCalls
			awk '{if ($7 == "PASS") print $0;}' sortedVariants.txt >> $outfile;
		#else
			cat sortedVariants.txt >> $outfile;
		#end if
		
	</command>
 
	<inputs>
		
		<!-- General Options -->
		<param name="type" type="select" multiple="true" display="checkboxes" label="Variant Types">
			<option value="DEL" selected="true">Deletions</option>
			<option value="DUP" selected="true">Tandem Duplications</option>
			<option value="INV" selected="true">Inversions</option>
			<option value="TRA" selected="true">Translocations</option>
		</param>
		<repeat name="repeatBam" title="Bam Alignment" min="1" default="1" >	
			<param format="bam" name="sortedBam" type="data" label="File" />		
		</repeat>
		<param name="excludeFile" type="data" optional="true" label="Chromosomes to Exclude"/> 
		<param name="filterCalls" type="boolean" value="false" label="Filter Poor Variant Calls"/>
		
		<!-- Paired End Options -->
		<param name="mapQual" type="integer" value="0" min="0" max="255" label="PE - Minimum Mapping Quality" />
		<param name="madCutoff" type="integer" value="9" min="0" max="255" label="PE - Insert Size Cutoff" />		

		<!-- SR Options -->
		<param format="fasta" name="genome" type="data" optional="true" label="SR - Genome Fasta File" />
		<param name="minFlank" type="integer" value="13" label="SR - Minimum Flanking Sequence" />

		<!-- Genotyping Options -->
		<param format="vcf" name="vcfgeno" type="data" optional="true" label="GT - Input VCF" />
		<param name="genoQual" type="integer" value="20" min="0" max="255" label="GT - Minimum Mapping Quality" />

	</inputs>
  
	<outputs>
		<data format="vcf" name="outfile" />
	</outputs>
  
	<help>

	</help>

</tool>