<tool id="tn93" name="TN93" version="@VERSION@">
    <description>compute distances between aligned sequences</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <requirements>
        <requirement type="package" version="@VERSION@">tn93</requirement>
    </requirements>
    <version_command><![CDATA[tn93 --version]]></version_command>
    <command detect_errors="exit_code"><![CDATA[
    tn93
    -o '$tn93_distances'
    #if str($options.advanced) == 'advanced':
        -t $options.threshold
        -a $options.ambigs
        -g $options.fraction
        -f csv
        -l $options.overlap
        #if $options.counts_in_name:
            -d '$options.counts_in_name'
        #end if
    #end if
    #if $second_fasta:
        -s '$second_fasta'
    #end if
    '$input_fasta'
    ]]></command>
    <inputs>
        <param name="input_fasta" type="data" format="fasta" label="Input in FASTA format" />
        <param name="second_fasta" argument="-s" type="data" format="fasta" optional="true"
            label="Additional input FASTA for pairwise comparisons" />
        <conditional name="options">
            <param label="Additional options" name="advanced" type="select">
                <option value="defaults">Use defaults</option>
                <option value="advanced">Specify additional parameters</option>
            </param>
            <when value="defaults" />
            <when value="advanced">
                <param name="threshold" type="float" value="0.015" label="Only report distances below this threshold" />
                <param name="ambigs" argument="-a" type="select" label="Strategy for ambiguous nucleotides">
                    <option value="handle">handle</option>
                    <option value="resolve">resolve</option>
                    <option value="average">average</option>
                    <option value="skip">skip</option>
                    <option value="gapmm">gapmm</option>
                </param>
                <param name="fraction" argument="-g" type="float" value="1.0"
                    label="Maximum tolerated fraction of ambiguous characters" />
                <param name="overlap" argument="-l" type="integer" value="100"
                    label="Only count pairs that overlap by at least N bases" />
                <param name="counts_in_name" argument="-d" type="text"
                    optional="true" label="Sequence name component immediately preceding a copy number" />
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data format="csv" name="tn93_distances" />
    </outputs>
    <tests>
        <test>
            <param name="input_fasta" value="tn93-in1.fa" />
            <param name="advanced" value="advanced" />
            <param name="threshold" value="0.35" />
            <output file="tn93-out1.csv" ftype="csv" name="tn93_distances" sort="True" />
        </test>
        <test>
            <param name="input_fasta" value="tn93-in2-alpha.fa" />
            <param name="second_fasta" value="tn93-in2-beta.fa" />
            <param name="advanced" value="advanced" />
            <param name="threshold" value="0.35" />
            <output file="tn93-out2.csv" ftype="csv" name="tn93_distances" sort="True" />
        </test>
    </tests>
    <help><![CDATA[
TN93
----

Compute pairwise distances between aligned
nucleotide sequences in sequential FASTA format using the
[Tamura Nei 93 distance](http://www.ncbi.nlm.nih.gov/pubmed/8336541).

NOTES
-----

All sequences must be aligned and have the same length. Only IUPAC characters are recognized (e.g. no ~). Sequence names can include copy number as in

  >seqname:10

':' can be replaced with another character using `-d`, and sequences that have no explicit copy number are assumed to be a single copy. Copy numbers
only affect histogram and mean calculations.
]]></help>
    <expand macro="citations">
        <citation type="doi">10.1093/oxfordjournals.molbev.a040023</citation>
    </expand>
</tool>