<?xml version="1.0"?>
<tool id="duplex" name="Du Novo: Make consensus reads" version="0.2">
  <description>from duplex sequencing alignments</description>
  <requirements>
    <requirement type="package" version="0.2">duplex</requirement>
    <requirement type="set_environment">DUPLEX_DIR</requirement>
  </requirements>
  <command detect_errors="exit_code"><![CDATA[
    python \$DUPLEX_DIR/duplex.py -r $min_reads -q $qual_thres -F $qual_format $input
    #if $keep_sscs:
      --sscs-file $sscs
    #end if
    > duplex.fa
    && awk -f \$DUPLEX_DIR/utils/outconv.awk -v target=1 duplex.fa > $output1
    && awk -f \$DUPLEX_DIR/utils/outconv.awk -v target=2 duplex.fa > $output2
  ]]>
  </command>
  <inputs>
    <param name="input" type="data" format="tabular" label="Aligned input reads" />
    <param name="min_reads" type="integer" value="3" min="1" label="Minimum reads per family" help="Single-strand families with fewer than this many reads will be skipped."/>
    <param name="qual_thres" type="integer" value="25" min="1" label="Minimum base quality" help="Bases with a PHRED score less than this will not be counted in the consensus making."/>
    <param name="qual_format" type="select" label="FASTQ format" help="Solexa should also work for Illumina 1.3+ and 1.5+, and Sanger should work for Illumina 1.8+">
      <option value="sanger" selected="true">Sanger (PHRED 0 = &quot;!&quot;)</option>
      <option value="solexa">Solexa (PHRED 0 = &quot;@&quot;)</option>
    </param>
    <param name="keep_sscs" type="boolean" truevalue="true" falsevalue="" label="Output single-strand consensus sequences" />
  </inputs>
  <outputs>
    <data name="output1" format="fasta" label="$tool.name on $on_string (mate 1)"/>
    <data name="output2" format="fasta" label="$tool.name on $on_string (mate 2)"/>
    <data name="sscs" format="fasta" label="$tool.name on $on_string (SSCS)">
      <filter>keep_sscs</filter>
    </data>
  </outputs>
  <tests>
    <test>
      <param name="input" value="families.msa.tsv"/>
      <output name="output1" file="families.cons_1.fa"/>
      <output name="output2" file="families.cons_2.fa"/>
    </test>
  </tests>
  <help>

**What it does**

This is for processing duplex sequencing data. It creates single-strand and duplex consensus reads from aligned read families.

-----

**Input**

This expects the output format of the "Align families" tool.

-----

**Output**

This will output final, duplex consensus reads in two FASTA files (first and second reads in the pairs). Optionally, you can save the single-strand reads too, in a separate FASTA file.

    </help>
</tool>