<?xml version="1.0" encoding="utf-8"?>
<tool id="snippy_core" name="snippy-core" version="@VERSION@">
    <description>
        Combine multiple Snippy outputs into a core SNP alignment
    </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
        #for $indir in $indirs
            #set $sample_name = os.path.splitext(os.path.basename(str($indir.name)))[0]
            mkdir '$sample_name' && tar -xf '$indir' -C '$sample_name' --strip-components=1 &&
        #end for
        #set snippy_dirs = " ".join(["'{0}'".format(os.path.splitext(os.path.basename(str($indir.name)))[0]) for $indir in $indirs])
        snippy-core
            --ref '$ref'
            ${snippy_dirs}
    ]]></command>

    <inputs>
        <param name="indirs" type="data" multiple="true" format="zip" label="Snippy input zipped dirs" help="Select all the snippy inputs for alignment" />
        <param name="ref" type="data" format="fasta,genbank" label="Reference File (either in fasta or genbank format)" help="Fasta or Genbank file to use as the reference" />
        <param name="outputs" type="select" multiple="true" display="checkboxes" label="Output selection">
            <option value="outaln" selected="True">A core SNP alignment in the fasta format</option>
            <option value="outfull" selected="False">A whole genome SNP alignment (includes invariant sites)</option>
            <option value="outtab" selected="False">Tab-separated columnar list of core SNP sites with alleles and annotations</option>
            <option value="outtxt" selected="False">Tab-separated columnar list of alignment/core-size statistics</option>
        </param>

    </inputs>

    <outputs>
        <data format="fasta" name="alignment_fasta" label="${tool.name} on ${on_string} core alignment fasta" from_work_dir="core.aln">
            <filter>outputs and 'outaln' in outputs</filter>
        </data>
        <data format="fasta" name="full_alignment_fasta" label="${tool.name} on ${on_string} full alignment fasta" from_work_dir="core.full.aln">
            <filter>outputs and 'outfull' in outputs</filter>
        </data>
        <data format="tabular" name="alignment_table" label="${tool.name} on ${on_string} core alignment table" from_work_dir="core.tab">
            <filter>outputs and 'outtab' in outputs</filter>
        </data>
        <data format="txt" name="alignment_summary" label="${tool.name} on ${on_string} core alignment summary" from_work_dir="core.txt">
            <filter>outputs and 'outtxt' in outputs</filter>
        </data>
    </outputs>

    <tests>
        <test><!-- Test #1 - test with 3 zipped directories -->
            <param name="indirs" value="a.tgz,b.tgz,c.tgz" />
            <param name="ref" value="reference.fasta" />
            <param name="outputs" value="outtxt" />
            <output name="alignment_summary" ftype="txt" file="a_b_c.core.txt" />
        </test>
    </tests>

    <help><![CDATA[
**snippy-core @VERSION@**

Combine multiple Snippy outputs into a core SNP alignment

If you call SNPs for multiple isolates from the same reference, you can produce an alignment of "core SNPs" which can be used to build a high-resolution phylogeny (ignoring possible recombination). A "core site" is a genomic position that is present in all the samples. A core site can have the same nucleotide in every sample ("monomorphic") or some samples can be different ("polymorphic" or "variant"). If we ignore the complications of "ins", "del" variant types, and just use variant sites, these are the "core SNP genome".


**Inputs:**

Multiple Snippy output directories. (At least 2 of)

**Options:**

    - noreference Exclude reference (default '0').

**Note:**

snippy **must** have been run with --cleanup False

    ]]></help>
    <expand macro="citations" />
</tool>