<?xml version="1.0"?>
<tool name="StringTie" id="stringtie" version="1.0.0">
    <description>RNA-Seq assembler</description>
    <requirements>
        <requirement type="package" version="0.97">stringtie</requirement>
    </requirements>
    <command>
<![CDATA[
stringtie "$input_bam"
-o "$output_gtf"
-p "\${GALAXY_SLOTS:-1}"
#if str($guide.use_guide) == 'yes':
    -G "$guide.guide_gff" $guide.input_estimation
    -C "$coverage"
#end if
#if str($option_set.options) == 'advanced':
    -f "$option_set.fraction"
    -m "$option_set.min_tlen"
    -a "$option_set.min_anchor_len"
    -j "$option_set.min_anchor_cov"
    -c "$option_set.min_bundle_cov"
    -s "$option_set.maxcov"
    -g "$option_set.bdist"    
    -M "$option_set.bundle_fraction" $option_set.sensitive $option_set.disable_trimming
#end if
]]>
</command>
    <inputs>
        <param name="input_bam" type="data" format="bam" label="BAM file to assemble" />
        <conditional name="guide">
            <param name="use_guide" type="select" label="Use GFF file to guide assembly">
                <option value="yes">Use GFF</option>
                <option value="no" selected="True">Do not use GFF</option>
            </param>
            <when value="no" />
            <when value="yes">
                <param name="guide_gff" type="data" format="gtf,gff3" label="Reference annotation to use for guiding the assembly process" />
                <param name="input_estimation" type="boolean" truevalue="-e" falsevalue="" label="Perform abundance estimation only of input transcripts" />
            </when>
        </conditional>
        <conditional name="option_set">
            <param name="options" type="select" label="Options">
                <option value="default" selected="True">Use defaults</option>
                <option value="advanced">Specify advanced options</option>
            </param>
            <when value="default" />
            <when value="advanced">
                <param name="disable_trimming" type="boolean" falsevalue="" truevalue="-t" label="Disable trimming of predicted transcripts based on coverage" />
                <param name="sensitive" type="boolean" truevalue="-S" falsevalue="" label="Increase sensitivity" />
                <param name="label" type="text" value="STRG" label="Name prefix for output transcripts" />
                <param name="fraction" type="float" value="0.15" min="0.0" max="1.0" label="Minimum isoform fraction" />
                <param name="min_tlen" type="integer" value="200" label="Minimum assembled transcript length" />
                <param name="min_anchor_len" type="integer" value="10" label="Minimum anchor length for junctions" />
                <param name="min_anchor_cov" type="integer" value="1" label="Minimum junction coverage" />
                <param name="min_bundle_cov" type="integer" value="2" label="Minimum bundle reads per bp coverage to consider for assembly" />
                <param name="maxcov" type="integer" value="1000000" label="Coverage saturation threshold" />
                <param name="bdist" type="integer" value="50" label="Gap between read mappings triggering a new bundle" />
                <param name="bundle_fraction" type="float" value="0.95" label="Fraction of bundle allowed to be covered by multi-hit reads" />
             </when>
        </conditional>
    </inputs>
    <outputs>
        <data name="output_gtf" format="gtf" label="${tool.name} on ${on_string}: Assembled transcripts"/>
        <data name="coverage" format="gff3" label="${tool.name} on ${on_string}: Coverage">
            <filter>guide['use_guide'] == "yes"</filter>
        </data>
    </outputs>
    <tests>
        <test>
            <param name="input_bam" value="stringtie_in1.bam" ftype="bam" />
            <param name="use_guide" value="no" />
            <param name="options" value="default" />
            <output name="output_gtf" file="stringtie_out1.gtf" ftype="gtf" />
        </test>
        <test>
            <param name="input_bam" value="stringtie_in1.bam" ftype="bam" />
            <param name="use_guide" value="no" />
            <param name="options" value="advanced" />
            <param name="fraction" value="0.17" />
            <output name="output_gtf" file="stringtie_out2.gtf" ftype="gtf" />
        </test>
    </tests>
    <help>
<![CDATA[
StringTie v0.97 usage::

   stringtie <input.bam> [-G <guide_gff>] [-l <label>] [-o <out_gff>] [-p <cpus>]
    [-v] [-a <min_anchor_len>] [-m <min_tlen>] [-j <min_anchor_cov>] [-n sens]
    [-C <coverage_file_name>] [-s <maxcov>] [-c <min_bundle_cov>] [-g <bdist>]

  Assemble RNA-Seq alignments into potential transcripts.

   Options:
   -G  reference annotation to use for guiding the assembly process (GTF/GFF3)
   -l  name prefix for output transcripts (default: STRG)
   -f  minimum isoform fraction (default: 0.15)
   -m  minimum assembled transcript length to report (default 200bp)
   -o  output file with the assembled transcripts (default: stdout)
   -a  minimum anchor length for junctions (default: 10)
   -j  minimum junction coverage (default: 1)
   -t  disable trimming of predicted transcripts based on coverage (default: trimming enabled)
   -c  minimum bundle reads per bp coverage to consider for assembly (default: 2)
   -s  coverage saturation threshold; further read alignments will be
       ignored in a region where a local coverage depth of <maxcov>
       is reached (default: 1,000,000);
   -v  verbose (log bundle processing details)
   -e  abundance estimation only of input transcripts (for -G option)
   -g  gap between read mappings triggering a new bundle (default: 50)
   -S  more sensitive run (default: no)
   -C  output file with all transcripts in reference that are fully
       covered by reads
   -M  fraction of bundle allowed to be covered by multi-hit reads (default:0.95)
   -p  number of threads (CPUs) to use (default: 1)

]]>
</help>
</tool>