<tool id="mira_4_0_mapping" name="MIRA v4.0 mapping" version="0.0.1">
    <description>Maps Sanger, Roche 454, Solexa/Illumina, Ion Torrent and PacBio reads</description>
    <requirements>
        <requirement type="python-module">Bio</requirement>
        <requirement type="binary">mira</requirement>
        <requirement type="package" version="4.0">MIRA</requirement>
    </requirements>
    <version_command interpreter="python">mira4.py -v</version_command>
    <command interpreter="python">
mira4.py $manifest $out_maf $out_fasta $out_log
    </command>
    <inputs>
        <param name="job_type" type="select" label="Assembly type">
            <option value="genome">Genome</option>
            <option value="est">EST (transcriptome)</option>
        </param>
        <param name="job_quality" type="select" label="Assembly quality grade">
            <option value="accurate">Accurate</option>
            <option value="draft">Draft</option>
        </param>
	<!-- TODO? Allow technology type for references? -->
	<!-- TODO? Allow strain settings for reference(s) and reads? -->
	<!-- TODO? Use a repeat to allow for multi-strain references? -->
        <!-- TODO? Add strain to the mapping read groups? -->
	<param name="references" type="data" format="fasta,fastq,mira" multiple="true" required="true" label="Backbone reference file(s)"
               help="Multiple files allowed, for example one FASTA file per chromosome or plasmid." />
        <param name="strain_setup" type="select" label="Strain configuration (reference vs reads)">
            <option value="default">Different strains - mapping reads onto a related reference ('StrainX' vs 'ReferenceStrain')</option>
            <option value="same">Same strain - mapping reads from same reference (all 'StrainX')</option>
        </param>
        <repeat name="read_group" title="Read Group" min="1">
            <param name="technology" type="select" label="Read technology">
                <option value="solexa">Solexa/Illumina</option>
                <option value="sanger">Sanger cappillary sequencing</option>
                <option value="454">Roche 454</option>
                <option value="iontor">Ion Torrent</option>
                <option value="pcbiolq">PacBio low quality (raw)</option>
                <option value="pcbiohq">PacBio high quality (corrected)</option>
                <option value="text">Synthetic reads (database entries, consensus sequences, artifical reads, etc)</option>
            </param>
            <param name="filenames" type="data" format="fastq,mira" multiple="true" required="true" label="Read file(s)"
                   help="Multiple files allowed, for example paired reads can be given as two files (MIRA looks at read names to identify pairs)." />
        </repeat>
    </inputs>
    <outputs>
        <data name="out_fasta" format="fasta" label="MIRA #if str($strain_setup)=='same' then 'same strain' else 'reference' # mapping contigs (FASTA)" />
        <data name="out_maf" format="mira" label="MIRA #if str($strain_setup)=='same' then 'same strain' else 'reference' # mapping assembly" />
        <data name="out_log" format="txt" label="MIRA #if str($strain_setup)=='same' then 'same strain' else 'reference' # mapping log" />
    </outputs>
    <configfiles>
        <configfile name="manifest">
project = MIRA
job = mapping,${job_type},${job_quality}
parameters = -GE:not=1 -NW:cmrnl -DI:trt=/tmp
## -GE:not is short for -GENERAL:number_of_threads and using one (1)
## can be useful for repeatability of assemblies and bug hunting.
##
## -NW:cmrnl is short for -NAG_AND_WARN:check_maxreadnamelength
## and without this MIRA aborts with read names over 40 characters
## due to limitations of some downstream tools.
##
## -DI:trt is short for -DIRECTORY:tmp_redirected_to and should
## point to a local hard drive (not something like NFS on network).

##This bar goes into the manifest as a comment line
#------------------------------------------------------------------------------

readgroup
is_reference
#if str($strain_setup)=="same"
strain = StrainX
#end if
#for $f in $references
##Must now map Galaxy datatypes to MIRA file types...
#if $f.ext.startswith("fastq")
##MIRA doesn't like fastqsanger etc, just plain old fastq:
data = fastq::$f
#elif $f.ext == "mira"
##We're calling *.maf the "mira" format in Galaxy (name space collision)
data = maf::$f
#elif $f.ext == "fasta"
##We're calling MIRA with the file type as "fna" as otherwise it wants quals
data = fna::$f
#else
##Currently don't expect anything else...
data = ${f.ext}::$f
#end if
#end for
#for $rg in $read_group

##This bar goes into the manifest as a comment line
#------------------------------------------------------------------------------

readgroup
technology = ${rg.technology}
#if str($strain_setup)=="same"
##This is perhaps redundant as MIRA defaults to StrainX for the reads:
strain = StrainX
#end if
##MIRA will accept multiple filenames on one data line, or multiple data lines
#for $f in $rg.filenames
##Must now map Galaxy datatypes to MIRA file types...
#if $f.ext.startswith("fastq")
##MIRA doesn't like fastqsanger etc, just plain old fastq:
data = fastq::$f
#elif $f.ext == "mira"
##We're calling *.maf the "mira" format in Galaxy (name space collision)
data = maf::$f
#else
##Currently don't expect anything else...
data = ${f.ext}::$f
#end if
#end for
#end for
        </configfile>
    </configfiles>
    <tests>
        <!-- Deliberately using default read_group.technology value "solexa"
             as then Galaxy's broken <repeat> handling in tests should work... -->
        <!-- Tests currently failing,
             TwillException: more than one form; you must select one (use 'fv') before submitting
        <test>
            <param name="job_type" value="genome" />
            <param name="job_quality" value="accurate" />
            <param name="references" value="tvc_contigs.fasta" ftype="fasta" />
            <param name="strain_setup" value="default" />
            <param name="filenames" value="tvc_mini.fastq" ftype="fastqsanger" />
            <output name="out_fasta" file="tvc_map_same_strain.fasta" ftype="fasta" />
        </test>
        <test>
            <param name="job_type" value="genome" />
            <param name="job_quality" value="accurate" />
            <param name="references" value="tvc_contigs.fasta" ftype="fasta" />
            <param name="strain_setup" value="same" />
            <param name="filenames" value="tvc_mini.fastq" ftype="fastqsanger" />
            <output name="out_fasta" file="tvc_map_ref_strain.fasta" ftype="fasta" />
        </test>
        -->
    </tests>
    <help>

**What it does**

Runs MIRA v4.0 in mapping mode, collects the output, and throws away all the temporary files.

MIRA is an open source assembly tool capable of handling sequence data from
a range of platforms (Sanger capillary, Solexa/Illumina, Roche 454, Ion Torrent
and also PacBio).

It is particularly suited to small genomes such as bacteria.

**Citation**

If you use this Galaxy tool in work leading to a scientific publication please
cite the following papers:

Peter J.A. Cock, Björn A. Grüning, Konrad Paszkiewicz and Leighton Pritchard (2013).
Galaxy tools and workflows for sequence analysis with applications
in molecular plant pathology. PeerJ 1:e167
http://dx.doi.org/10.7717/peerj.167

Bastien Chevreux, Thomas Wetter and Sándor Suhai (1999).
Genome Sequence Assembly Using Trace Signals and Additional Sequence Information.
Computer Science and Biology: Proceedings of the German Conference on Bioinformatics (GCB) 99, pp. 45-56.
http://www.bioinfo.de/isb/gcb99/talks/chevreux/main.html

This wrapper is available to install into other Galaxy Instances via the Galaxy
Tool Shed at http://toolshed.g2.bx.psu.edu/view/peterjc/mira4_assembler
    </help>
</tool>