<tool id="yaha" name="yaha" version="0.1.83">
    <description>fast and flexible long-read alignment with optimal breakpoint detection</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <requirements>
        <requirement type="package" version="0.1.83">yaha</requirement>
        <requirement type="package" version="0.6.5">sambamba</requirement>
    </requirements>
    <version_command><![CDATA[yaha 2>&1| head -n1]]></version_command>
    <command detect_errors="aggressive"><![CDATA[
        #if $reference_genome.reference_genome_source == "history" or "reference":
            #if $reference_genome.reference_genome_source == "history":
                ln -f -s "$reference_genome.history_item" genome.fa &&
            #elif $reference_genome.reference_genome_source == "reference":
                ln -f -s "$reference_genome.fasta_item.fields.path" genome.fa &&
            #end if
            yaha -g genome.fa -H $reference_genome.maxHitsIndex -L $reference_genome.wordLenIndex -S $reference_genome.skipDistanceIndex &&
            #set maxHitsIndex = "%s%s" % ("0" * (5 - len(str($reference_genome.maxHitsIndex))), $reference_genome.maxHitsIndex)
            #set skipdist = $reference_genome.skipDistanceIndex if len(str($reference_genome.skipDistanceIndex)) > 1 else "0%s" % $reference_genome.skipDistanceIndex
            #set wordLenIndex = $reference_genome.wordLenIndex if len(str($reference_genome.wordLenIndex)) > 1 else "0%s" % $reference_genome.wordLenIndex
            #set index_path = "genome.X%s_%s_%sS" % ($wordLenIndex, $skipdist, $maxHitsIndex)
        #else:
            #pass  ## augment with pre-built index if considered useful
        #end if
        yaha
        -x '$index_path'
        #if str($readtype.single_or_paired) == "se":
            -q '$reatype.input_single'
        #else:
            -q <(cat '$readtype.input_paired.forward' '$readtype.input_paired.reverse')
        #end if
        $outformat stdout
        -t \${GALAXY_SLOTS:-1}
        -BW $BW
        -G $G
        -H $H
        -M $M
        -MD $MD
        -P $P
        -X $X
        #if $ags.use_ags == "yes":
            -AGS Y
            -GEC $ags.GEC
            -GOC $ags.GOC
            -MS $ags.MS
            -RC $ags.RC
        #else:
            -AGS N
        #end if
        #if $oqc.use_oqc == "yes":
             -OQC Y
             -BP $oqc.BP
             -MGDP $oqc.MGDP
             -MNO $oqc.MNO
        #else:
             -OQC N
        #end if
        #if $fbs.use_fbs == "yes":
             -FBS Y
             -PRL $fbs.PRL
             -PSS $fbs.PSS
        #else:
             -FBS N
        #end if
        | sambamba view -S -f bam /dev/stdin | sambamba sort -o '$alignment' -l 6 -t \${GALAXY_SLOTS:-1} /dev/stdin
        ]]></command>
    <inputs>
        <conditional name="readtype">
            <param name="single_or_paired" type="select" label="Single-end or paired-end reads?" help="Note: yaha does not take mate-pair information into account. If you select pe_collection, forward and reverse reads will be merged.">
                <option value="se" selected="true">Single-end</option>
                <option value="pe_collection">Paired-end (as collection)</option>
            </param>
            <when value="se">
                <param format="fastqsanger,fastqsanger.gz" name="input_single" type="data" label="Single-end FASTQ reads" help="(-q)" />
            </when>
            <when value="pe_collection">
                <param format="fastqsanger,fastqsanger.gz" name="input_paired" type="data_collection" collection_type="paired" label="Paired-end FASTQ reads as paired collection" help="(-q)"/>
            </when>
        </conditional>
        <conditional name="reference_genome">
            <param help="Built-in references were created using default options" label="Source for the reference genome to align against" name="reference_genome_source" type="select">
                <!-- Write a datamanager if prebuilt genomes are important
                <option selected="True" value="indexed">Use a built-in index</option>
                -->
                <option value="history">Use a genome from history to build an index</option>
                <option value="reference">Use a built-in genome to build an index</option>
            </param>
            <!--
            <when value="indexed">
                <param help="If your genome of interest is not listed, contact the Galaxy team" label="Select a reference genome" name="index" type="select">
                    <options from_data_table="yaha_indexes">
                        <filter column="2" type="sort_by" />
                        <validator message="No genomes are available for the selected input dataset" type="no_options" />
                    </options>
                </param>
            </when>
            -->
            <when value="history">
                <param format="fasta" label="Select the reference genome" name="history_item" type="data" />
                <expand macro="index_parameter"/>
            </when>
            <when value="reference">
                <param label="Select a reference genome" name="fasta_item" type="select">
					<options from_data_table="all_fasta">
                        <filter column="2" type="sort_by"/>
					</options>
				</param>
                <expand macro="index_parameter"/>
            </when>
        </conditional>
        <param name="outformat" type="select" label="Produce alignment with softclipping?">
            <option value="-osh">Produce alignment with hard clipping</option>
            <option value="-oss">Produce alignment with soft clipping</option>
        </param>
        <param type="integer" argument="-BW" value="5" min="0" label="BandWidth" help="band size on each side of the diagonal of banded Smith Waterman" />
        <param type="integer" argument="-G" value="50" min="0" label="maxGap" help="maximum indel size allowed with a single alignment" />
        <param type="integer" argument="-H" value="650" min="1" max="65525" label="maxHits" help="maximum times a seed is in the reference before it is ignored as too repetitive. To take advantage of k-mer sampling, use the same value of maxHits during index creation and alignment." />
        <param type="integer" argument="-M" value="25" min="0" label="minMatch" help="minimum number of bases in seeds to start an alignment" />
        <param type="integer" argument="-MD" value="50" min="0" label="MaxDesert" help="maximum number of contiguous bases without a seed before alignmment is split" />
        <param type="float" argument="-P" value="0.9" min="0" label="minPercent-identity" help="minimum matching/alignment-length for a query to be included in output" />
        <param type="integer" argument="-X" value="25" min="0" label="Xdropoff" help="maximum score dropoff before terminating alignment extensions" />
        <conditional name="ags">
            <param name="use_ags" type="select" label="Use Affine Gap Scoring?">
                <option value="yes" selected="True">Yes</option>
                <option value="no">No</option>
            </param>
            <when value="yes">
				<param argument="-GEC" type="integer" value="2" min="0" label="GapExtensionCost" help="cost for extending a gap (indel)"/>
				<param argument="-GOC" type="integer" value="5" min="0" label="GapOpenCost" help="cost for starting a new gap (indel)"/>
                <param argument="-MS" type="integer" value="1" min="0" label="MatchScore" help="score added for each matching base"/>
                <param argument="-RC" type="integer" value="3" min="0" label="ReplacementCost" help="score subtracted for each mismatched base"/>
            </when>
        <when value="no">
        </when>
    </conditional>
        <conditional name="oqc">
            <param name="use_oqc" type="select" label="Use Optimal Query Coverage Algorithm?" help="">
                <option value="yes" selected="True">Yes (Find a set of alignments are found that optimally cover the query, using the remaining options)</option>
                <option value="no">No (Output all alignments meeting above criteria)</option>
            </param>
            <when value="yes">
                <param argument="-BP" type="integer" value="5" min="0" label="BreakpointPenalty" help="penalty for inserting a breakpoint in split-read alignment"/>
                <param argument="-MGDP" type="integer" value="5" min="0" label="MaxGenomicDistancePenalty"/>
                <param argument="-MNO" type="integer" value="25" min="0" label="MinNonOverlap" help="minimum number of unshared bases required in each split alignment"/>
            </when>
        <when value="no">
        </when>
    </conditional>
        <conditional name="fbs">
            <param name="use_fbs" type="select" label="Use Filter By Similarity Algorithm?" help="">
                <option value="yes" selected="False">Yes (Output alignments similar to best alignment found using OQC.)</option>
                <option value="no">No</option>
            </param>
            <when value="yes">
                <param argument="-PRL" type="float" value="0.9" min="0" max="1" label="PercentReciprocalLength" help="minimum ratio of overlapping length between similar alignment"/>
                <param argument="-PSS" type="float" value="0.9" min="0" max="1" label="PercentSimilarScore" help="minimum ratio of scores between similar alignments"/>
            </when>
        <when value="no">
        </when>
    </conditional>
    </inputs>
    <outputs>
        <data name="alignment" format="bam" />
    </outputs>
    <tests>
        <test>
            <param name="q" value="input.fastq" ftype="fastqsanger"/>
            <param name="reference_genome_source" value="history"/>
            <param name="history_item" value="phiX.fa" ftype="fasta"/>
            <output name="alignment" value="alignment.bam" ftype="bam"/>
        </test>
    </tests>
    <help><![CDATA[
Summary
-------

*yaha* is an open source, flexible, sensitive and accurate DNA aligner
designed for single-end reads. It supports three major modes of
operation:

1. The default “Optimal Query Coverage” (**-OQC**) mode reports the best
   set of alignments that cover the length of each query.
2. Using “Filter By Similarity” (**-FBS**), along with the best set of
   alignments, *yaha* will also output alignments that are highly
   similar to an alignment in the best set.
3. Finally, *yaha* can output all the alignments found for each query.

The **-OQC** and **-FBS** modes are specifically tuned to form split
read mappings that can be used to accurately identify structural
variation events (deletions, duplications, insertions or inversions)
between the subject query and the reference genome.

Usage
-----

**OPTIONS:** Default values enclosed in square brackets []

::

    Input/Output Options:
    -g    FILE input genome file to use during index creation (FASTA or nib2)
    -q    FILE input file of sequence reads to align (FASTA or FASTQ) [STDIN]
    -osh  FILE output file for alignment output in SAM format with hard clipping(default) [STDOUT]
    -oss  FILE output file for alignment output in SAM format with soft clipping [STDOUT]
    -x    FILE reference index file to use during alignment
    NOTE: At most one of -osh or -oss should be specified.

    Index Creation Options:
    -H    INT  maxHits: During index creation, seeds occuring more than maxHits times will be sampled [65565]
    -L    INT  seedLength: Length of seed to use.  During alignment, seed length is taken from index file [15]
    -S    INT  Skipdistance: Number of bases to skip ahead before forming next seed [1]

    General Alignment Options:
    -BW   INT  BandWidth: band size on each side of the diagonal of banded Smith Waterman [5]
    -G    INT  maxGap: maximum indel size allowed with a single alignment [50]
    -H    INT  maxHits: maximum times a seed is in the reference before it is ignored as too repetitive [650]
    -M    INT  minMatch: minimum number of bases in seeds to start an alignment [25]
    -MD   INT  MaxDesert: maximum number of contiguous bases without a seed before alignmment is split [50]
    -P    REAL minPercent-identity: minimum matching/alignment-length for a query to be included in output [0.9]
    -X    INT  Xdropoff: maximum score dropoff before terminating alignment extensions [25]
    -t    INT  numThreads: number of threads used to parallel process reads [1]

    Affine Gap Scoring Options:
    If -AGS is off, a simple edit distance calculation is done.
    If on, the remaining options are used:
    -AGS  BOOL (Y|N) controls use of Affine Gap Scoring [Y].
    -GEC  INT  GapExtensionCost: cost for extending a gap (indel) [2]
    -GOC  INT  GapOpenCost: cost for starting a new gap (indel) [5]
    -MS   INT  MatchScore: score added for each matching base [1]
    -RC   INT  ReplacementCost: score subtracted for each mismatched base [3]

    Optimal Query Coverage Options:
    If -OQC if off, all alignments meeting above criteria are output.
    If -OQC is on, a set of alignments are found that optimally cover the query, using the remaining options.
    -OQC  BOOL (Y|N) controls use of the Optimal Query Coverage Algorithm.
    -BP   INT BreakpointPenalty: penalty for inserting a breakpoint in split-read alignment [5]
    -MGDP INT MaxGenomicDistancePenalty (5)]
    -MNO  INT MinNonOverlap: minimum number of unshared bases required in each split alignment [minMatch]
    NOTE: The total cost of adding a breakpoint in a split-read mapping is:
      BP*MIN(MGDP, Log10(genomic distance between reference loci))

    Filter By Similarity Options:
    If -FBS is on, the remaining options are used.  An alignemnt must satisfy BOTH criteria to be "similar".
    -FBS  BOOL (Y|N) controls output of alignments similar to best alignment found using OQC.
    -PRL  REAL PercentReciprocalLength: minimum ratio of overlapping length between similar alignemnt [0.9]
    -PSS  REAL PercentSimilarScore: minimum ratio of scores between similar alignments [0.9]

See the `User Guide <https://www.dropbox.com/s/7j758vpbaskcq20/YAHA_User_Guide.0.1.83.pdf?dl=0>`__
for more details on all options and their usage.

| **Written by:** Greg Faust (gf4ea@virginia.edu)
| `Ira Hall Lab, University of
  Virginia <http://faculty.virginia.edu/irahall/>`__

| **Please cite:**
| `Faust G.G. and Hall I.M., "*YAHA*: fast and flexible long-read
  alignment with optimal breakpoint detection," *Bioinformatics* Oct.
  2012; **28**\ (19):
  2417-2424. <http://bioinformatics.oxfordjournals.org/content/28/19/2417>`__
        ]]></help>
    <citations>
        <citation type="doi">10.1093/bioinformatics/bts456</citation>
    </citations>
</tool>