<tool id="redup" name="Redup" version="1.0">
<description>Remove exact duplicate reads from paired-end fastq files</description>
<command interpreter="perl"> 
  ## Need to handle file names, probably should fix redup.pl
  redup.pl 
  #if $opt_n:
    -n $opt_n 
  #end if
  $fastq1_in $fastq2_in  $unique1_out $unique2_out
  #if $opt_n:
    > $duplicates
  #end if
</command>
<inputs>
  <param name="fastq1_in" type="data" format="fastq" label="Fastq Input 1"/>
  <param name="fastq2_in" type="data" format="fastq" label="Fastq Input 2"/>
  <param name="opt_n" type="integer" value="20" optional="true" label="Number of most duplicated sequences printed out. (default 20)"> 
    <validator type="in_range" message="Value can not be negative" min="0"/>
  </param>
</inputs>
<stdio>
  <exit_code range="1:" level="fatal"/>
</stdio>
<outputs>
  <data format_source="fastq1_in" name="unique1_out" label="${tool.name} on ${on_string}: fastq1.unique" />
  <data format_source="fastq2_in" name="unique2_out" label="${tool.name} on ${on_string}: fastq2.unique" />
  <data format="fasta" name="duplicates" label="${tool.name} on ${on_string}: top duplicates" >
    <filter>opt_n != None</filter>
  </data>
</outputs>
<tests>
  <test>
    <param name="fastq1_in" ftype="fastq" value="input1.fastq" />
    <param name="fastq2_in" ftype="fastq" value="input2.fastq" />
    <param name="opt_n" value="20" />
    <output name="unique1_out" file="output1.fastq" />
    <output name="unique2_out" file="output2.fastq" />
    <output name="duplicates" file="duplicates.fasta" />
  </test>
</tests>

<help>
This script removes duplicate paired-end reads from the input files sample1_R1.fastq and sample1_R2.fastq and prints out unique reads to the files sample1_R1.fastq.unique and sample2_R2.fastq.unique. Reads must have the exact same sequence to be called duplicates, quality scores are ignored. The top N (default 20) most duplicated sequences are printed out in fasta format, making it convenient for using BLAST to identify them.
</help>
</tool>