<tool id="tatajuba" name="tatajuba" version="@VERSION@">
    <description> Finding distribution of homopolymeric tracts</description>
    <macros>
        <token name="@VERSION@">1.0.4</token>
    </macros>
    <requirements>
        <requirement type="package" version="@VERSION@">tatajuba</requirement>
    </requirements>
    <version_command>tatajuba --version</version_command>
    <command detect_errors="exit_code"><![CDATA[
        #if $reads.sr_type == "paired"
            ln -s '$reads.R1' read1.fastq.gz &&
            ln -s '$reads.R2' read2.fastq.gz &&
        #end if
        #if $reads.sr_type == "collection"
            ln -s '$reads.input1.forward' read1.fastq.gz &&
            ln -s '$reads.input1.reverse' read2.fastq.gz &&
        #end if
        ln -s '$gff' 'ref.gff' &&
        ln -s '$fasta' 'ref.fasta' &&
        tatajuba
        $keep_bias
        $vcf
        --kmer=$kmer
        --minsize=$minsize
        --minreads=$minreads
        --maxdist=$maxdist
        #if $leven > 0
        --level $leven
        #end if
        --nthreads=\${GALAXY_SLOTS:-4}
        --gff=ref.gff
        --fasta=ref.fasta
        *.fastq.gz
        --outdir=output
    ]]>    </command>

    <inputs>
        <conditional name="reads">
            <param name="sr_type" type="select" label="Input reads type or collection" help="Select 'paired end' for a single library or 'collection' for a paired end collection">
                <option value="paired" selected="true">Paired End</option>
                <option value="collection">Paired Collection</option>
            </param>
            <when value="paired">
                <param name="R1" type="data" format="fastqsanger,fastqsanger.gz,fastqsanger.bz2" label="Forward reads (R1)" help="The file of forward reads in FASTQ format"/>
                <param name="R2" type="data" format="fastqsanger,fastqsanger.gz,fastqsanger.bz2" label="Reverse reads (R2)" help="The file of reverse reads in FASTQ format"/>
            </when>
            <when value="collection">
                <param name="input1" format="fastqsanger,fastqsanger.gz,fastqsanger.bz2" type="data_collection" collection_type="paired" label="Paired collection" help="See help section for an explanation of dataset collections"/>
            </when>
        </conditional>
        <param name="gff" type="data" format="gff" label="Reference in GFF format"/>
        <param name="fasta" type="data" format="fasta,fasta.gz" label="Reference in fasta format"/>
        <param argument="kmer" type="integer" value="25" min="2" max="32" label="Kmer size flanking each side of homopolymer" />
        <param argument="minsize" type="integer" value="4" min="1" max="32" label="Fraction value of the read depth to be considered invalid" />
        <param argument="minreads" type="integer" value="5" min="1" label="Minimum number of reads for tract+context to be considered" />
        <param argument="maxdist" type="integer" value="1" min="1" label="Maximum distance between kmers of a flanking region to merge them into one context" />
        <param argument="leven" type="integer" value="0" min="0" label="Levenshtein distance between flanking regions to merge them into one context" />
        <param argument="vcf" type="boolean" truevalue="--vcf" falsevalue="" checked="false" label="generate VCF files for each sample, around the HT regions" />
        <param argument="keep_bias" type="boolean" truevalue="--keep_bias" falsevalue="" checked="false" label="Keep biased tracts" />
    </inputs>
    <outputs>
        <data name="variable_tracts" format="tabular" from_work_dir="output/variable_tracts.bed" label="${tool.name} on ${on_string} BED file with tract locations" />
        <data name="tract_list" format="tabular" from_work_dir="output/tract_list.tsv" label="${tool.name} on ${on_string} list of all HTs found" />
        <data name="selected_tracts_unknown" format="tabular" from_work_dir="output/selected_tracts_unknown.tsv" label="${tool.name} on ${on_string} debug file with difference stats per tract for tracts outside annotated regions" />
        <data name="selected_tracts_annotated" format="tabular" from_work_dir="output/selected_tracts_annotated.tsv" label="${tool.name} on ${on_string} debug file with difference stats per tract for tracts in annotated regions" />
        <data name="per_sample_proportional_coverage" format="tabular" from_work_dir="output/per_sample_proportional_coverage.tsv" label="${tool.name} on ${on_string} feature matrix of proportional coverage depth of HT" />
        <data name="per_sample_proportional_coverage" format="tabular" from_work_dir="output/per_sample_modal_frequency.tsv" label="${tool.name} on ${on_string} feature matrix with histogram bar length of modal tract length" />
        <data name="per_sample_average_length" format="tabular" from_work_dir="output/per_sample_average_length.tsv" label="${tool.name} on ${on_string} feature matrix with average HT length" />
        <data name="vcf1" format="vcf" from_work_dir="output/1.vcf.gz" label="${tool.name} on ${on_string} VCF-1" />
        <data name="vcf2" format="vcf" from_work_dir="output/2.vcf.gz" label="${tool.name} on ${on_string} VCF-2" />
    </outputs>
    <tests>
        <test name="test-paired-end">
            <param name="sr_type" value="paired" />
            <param name="R1" value="r1.fastq.gz" ftype="fastqsanger.gz" />
            <param name="R2" value="r2.fastq.gz" ftype="fastqsanger.gz" />
            <param name="gff" value="GCF_000009085.1_ASM908v1_genomic.gff" ftype="gff" />
            <param name="fasta" value="GCF_000009085.1_ASM908v1_genomic.fna" ftype="fasta" />
            <output name="per_sample_average_length" file="output/per_sample_average_length.tsv" ftype="tabular" compare="contains" />
        </test>
    </tests>
    <help><![CDATA[
    ]]>    </help>
</tool>