<tool id="breseq" name="breseq" version="@PACKAGE_VERSION@+@GALAXY_VERSION@">

    <description>find mutations in haploid microbial genomes</description>

    <macros>
        <import>macros.xml</import>
    </macros>

    <requirements>
        <requirement type="package" version="0.35.5">breseq</requirement>
        <requirement type="package" version="1.32">tar</requirement>
    </requirements>

    <version_command>breseq --version</version_command>

    <command detect_errors="aggressive">
        <![CDATA[
            #set $ref_opts = ""
            #for $i, $r in enumerate( $references ):
                #if str($references[$i].reference.source) == "history":
                    #for $ref in $references[$i].reference.own_genome:
                        #if $ref
                            #set $ref_opts = $ref_opts + " --reference '" + str($ref) + "'"
                        #end if
                    #end for
                #else:
                    #set $ref_opts = $ref_opts + " --reference '" + $references[$i].reference.fixed_genome.fields.path + "'"
                #end if
            #end for

            #if str($run.mode) == 'detect'
                breseq

                --num-processors \${GALAXY_SLOTS:-4}

                -o results

                $ref_opts

                #for $s in $run.fastqs:
                    ${s}
                #end for

                #if $run.name
                    --name '$run.name'
                #end if

                $run.polymorphism_prediction
                $run.predict_junctions

                #if 'gd' in str($run.output_options.formats).split(','):
                    && cp results/output/output.gd '$output'
                #end if

                #if 'html' in str($run.output_options.formats).split(','):
                    && cp results/output/index.html '$report'
                    && mkdir $report.extra_files_path
                    && cp -R results/output/* $report.extra_files_path
                #end if

                #if 'zip' in str($run.output_options.formats).split(','):
                    && tar -zcf '$zip_output' results
                #end if

                #if 'log' in str($run.output_options.formats).split(','):
                    && cp results/output/log.txt '$log'
                #end if
            #else
                #set $first = 1
                #for $o in str($run.output_options.formats).split(','):

                    #if $first == 0
                        &&
                    #end if
                    #set $first = 0

                    gdtools ANNOTATE

                    --format '$o'

                    -o
                    #if $o == 'html':
                        '$annreport'
                    #else if $o == 'gd':
                        '$genomediff'
                    #else if $o == 'tsv':
                        '$tabdelim'
                    #else if $o == 'phylip':
                        '$phylipout'
                    #else if $o == 'json':
                        '$jsonout'
                    #end if

                    $ref_opts

                    #for $s in $run.gds:
                        ${s}
                    #end for
                #end for
            #end if
        ]]>
    </command>

    <inputs>
        <repeat name="references" title="Reference Genome" min="1">
            <conditional name="reference">
                <param name="source" type="select" label="Reference source" >
                    <option value="builtin">built-in</option>
                    <option value="history" selected="true">history</option>
                </param>
                <when value="builtin">
                    <param name="fixed_genome" argument="--reference" type="select" optional="false" label="Galaxy Built-in Reference(s)">
                        <options from_data_table="genbank_files">
                            <filter type="sort_by" column="3"/>
                            <validator type="no_options" message="No built-in genbank records have been configured"/>
                        </options>
                    </param>
                </when>
                <when value="history">
                    <param name="own_genome" argument="--reference" type="data" format="fasta,genbank" multiple="true" optional="false" label="Fasta or Genbank Reference(s)" />
                </when>
            </conditional>
        </repeat>

        <conditional name="run">
            <param name="mode" type="select" label="Run Mode" help="Detect, annotate, or compare variants.">
                <option value="detect" selected="true">Detect</option>
                <option value="annotate">Annotate</option>
                <option value="compare">Compare</option>
            </param>
            <when value="detect">

                <param name="fastqs" type="data" format="fastq,fastq.gz" multiple="true" label="Fastq Read Files" />

                <param argument="--polymorphism-prediction" name="polymorphism_prediction" type="select" label="Detection Mode" help="**Polymorphism mode**: Detect variants with frequencies between 0% and 100% if a mixture model is well-supported by the read alignment evidence. Use to analyze a mixed population of genomes evolved from a common ancestor. **Consensus mode**: Detect variants present in 100% of the sample. Use when re-sequencing a clonal haploid genome. This mode is the default.">
                    <option value="" selected="true">Consensus</option>
                    <option value="--polymorphism-prediction">Polymorphism</option>
                </param>

                <param name="name" argument="--name" type="text" value="" label="Analysis Name" help="Human-readable name of the analysis run for output (DEFAULT=none)." />

                <param name="predict_junctions" type="boolean" truevalue="" falsevalue="--no-junction-prediction" checked="true" label="Predict Junctions" help="Predict new sequence junctions (default).  --no-junction-prediction is supplied if 'No' is selected.  Otherwise, there is no flag." />

                <section name="output_options" title="Output Options" expanded="false">
                    <param name="formats" type="select" multiple="true" optional="false" display="checkboxes" label="Output Formats">
                        <option value="gd" selected="true">Variants (GenomeDiff)</option>
                        <option value="html">Variant Report (Webpage)</option>
                        <option value="zip">All Variant Results (Gzip)</option>
                        <option value="log">Log (Text)</option>
                    </param>
                </section>

            </when>
            <when value="annotate">

                <param name="gds" type="data" format="tabular" multiple="true" optional="false" label="GenomeDiff (gd) Files" help="Files as produced by breseq" />

                <expand macro="annotate_format_opts">
                    <option value="gd" selected="true">Annotated Variants (GenomeDiff)</option>
                </expand>

            </when>
            <when value="compare">

                <param name="gds" type="data" format="tabular" multiple="true" optional="false" label="GenomeDiff (gd) Files" help="Files as produced by breseq" min="2" />

                <expand macro="annotate_format_opts">
                    <option value="phylip" selected="true">Variant Comparison (Phylip)</option>
                    <option value="gd">Annotated Variants (GenomeDiff)</option>
                </expand>

            </when>
        </conditional>

    </inputs>

    <outputs>
        <data format="html" name="report" label="${tool.name} on ${on_string}: Variants (Webpage)">
            <filter>run['mode'] == 'detect' and 'html' in run['output_options']['formats']</filter>
        </data>
        <data format="html" name="annreport" label="${tool.name} on ${on_string}: Annotated Variants Report (Webpage)">
            <filter>run['mode'] != 'detect' and 'html' in run['output_options']['formats']</filter>
        </data>

        <data format="tabular" name="output" label="${tool.name} on ${on_string}: Variants (GenomeDiff)">
            <filter>run['mode'] == 'detect' and 'gd' in run['output_options']['formats']</filter>
        </data>
        <data format="tabular" name="genomediff" label="${tool.name} on ${on_string}: Annotated Variants (GenomeDiff)">
            <filter>run['mode'] != 'detect' and 'gd' in run['output_options']['formats']</filter>
        </data>

        <data format="zip" name="zip_output" label="${tool.name} on ${on_string}: All Variant Results (Gzip)">
            <filter>'zip' in run['output_options']['formats']</filter>
        </data>
        <data format="txt" name="log" label="${tool.name} on ${on_string}: Breseq Log">
            <filter>'log' in run['output_options']['formats']</filter>
        </data>
        <data format="tabular" name="tabdelim" label="${tool.name} on ${on_string}: Annotated Variants (Tabular)">
            <filter>'tsv' in run['output_options']['formats']</filter>
        </data>
        <data format="phylip" name="phylipout" label="${tool.name} on ${on_string}: Variant Comparison (Phylip)">
            <filter>'phylip' in run['output_options']['formats']</filter>
        </data>
        <data format="txt" name="jsonout" label="${tool.name} on ${on_string}: Annotated Variants (JSON)">
            <filter>'json' in run['output_options']['formats']</filter>
        </data>
    </outputs>

    <tests>
        <test>
            <repeat name="references">
                <conditional name="reference">
                    <param name="source" value="history" />
                    <param name="own_genome" value="lambda.gbk" />
                </conditional>
            </repeat>
            <conditional name="run">
                <param name="mode" value="detect" />
                <param name="fastqs" value="lambda.short_sequence_repeats.fastq.gz" ftype="fastq.gz" />
                <param name="polymorphism_prediction" value="" />
                <param name="name" value="smallest" />
                <param name="predict_junctions" value="" />
                <section name="output_options">
                    <param name="formats" value="html,log,gd,zip" />
                </section>
            </conditional>

            <output name="report" file="report.html" compare="sim_size" delta="100" />
            <output name="log" file="log.txt" lines_diff="4">
                <assert_contents>
                    <has_text text="breseq --num-processors" />
                </assert_contents>
            </output>
            <output name="output" file="gdout.txt" lines_diff="8" />
            <output name="zip_output">
                <assert_contents>
                    <has_archive_member path="results/output/output.gd" />
                </assert_contents>
            </output>
        </test>
        <test>
            <repeat name="references">
                <conditional name="reference">
                    <param name="source" value="builtin" />
                    <param name="fixed_genome" value="lambda1" />
                </conditional>
            </repeat>
            <conditional name="run">
                <param name="mode" value="detect" />
                <param name="fastqs" value="lambda.short_sequence_repeats.fastq.gz" ftype="fastq.gz"/>
                <param name="polymorphism_prediction" value="" />
                <param name="name" value="smallest" />
                <param name="predict_junctions" value="" />
                <section name="output_options">
                    <param name="formats" value="gd" />
                </section>
            </conditional>

            <output name="output" file="gdout.txt" lines_diff="8" />
        </test>
        <test>
            <repeat name="references">
                <conditional name="reference">
                    <param name="source" value="history" />
                    <param name="own_genome" value="lambda.gbk" />
                </conditional>
            </repeat>
            <conditional name="run">
                <param name="mode" value="annotate" />
                <param name="gds" value="gdout.txt" />
                <section name="output_options">
                    <param name="formats" value="html" />
                </section>
            </conditional>

            <output name="annreport" file="gdtoolsout.html" compare="sim_size" delta="100" />
        </test>
    </tests>

    <help>
        <![CDATA[
**Detect Variants**

breseq (pronounced: \\brēz-ˈsēk\\ or breeze-seq) is a computational pipeline for
the analysis of short-read re-sequencing data (e.g. Illumina, 454, IonTorrent,
etc.). It uses reference-based alignment approaches to predict mutations in a
sample relative to an already sequenced genome. breseq is intended for microbial
genomes (<10 Mb) and re-sequenced samples that are only slightly diverged from
the reference sequence (<1 mutation per 1000 bp).

breseq's primary advantages over other software programs are that it can:

- Accurately predict new sequence junctions, such as those associated with mobile element insertions.
- Integrate multiple sources of evidence for genetic changes into mutation predictions.
- Produce annotated output describing biologically relevant mutational events.

breseq was initially developed to analyze data from the Lenski long-term
evolution experiment with `E. coli`_. References: barrick2009a_ barrick2009b_.

.. _`E. coli`: http://myxo.css.msu.edu/ecoli/
.. _barrick2009a: http://barricklab.org/twiki/pub/Lab/ToolsBacterialGenomeResequencing/documentation/references.html#barrick2009a
.. _barrick2009b: http://barricklab.org/twiki/pub/Lab/ToolsBacterialGenomeResequencing/documentation/references.html#barrick2009b

However, breseq may be generally useful to researchers who are:

- Tracking mutations over time in microbial evolution experiments.
- Checking strains for unwanted second-site mutations after genetic manipulations.
- Identifying mutations that occur during strain improvement or after long-term culture of engineered strains.
- Discovering what mutations arise in pathogens during infection or cause antibiotic resistance.


*Inputs*

Breseq accepts files in FASTQ format. It does not take pair-end information into
account.

You can either run in clonal (consensus) mode or search for polymorphisms in a
population.

You can also select an external sequence (eg. a transposon) to detect for
insertions or horizontal transfer.


*Outputs*

Breseq outputs a number of files. These are all condensed in a single zipped
file.

It contains output files with the final results, accessible through
``output/index.html``

It also contains data files with accessory data, including:

- ``data/reference.fasta`` (file with reference genome: can be used in eg. IGV browser)
- ``data/reference.gff`` (file with genomic annotations: can be used in eg. IGV browser)
- ``data/areference.bam`` (file with read alignments: can be used in eg. IGV browser)
- ``data/unmatched.*`` (files with read that failed to align: can be used to build an assembly or to eg. blast against NCBI)


----

**Annotate Variants**

Annotate a GenomeDiff file (generated by breseq) with information about
mutations (what genes they affect, amino acid substitutions, etc.) If multiple
input files are provided, then also COMPARE the frequencies for identical
mutations across samples.
        ]]>
    </help>

    <citations>
        <citation type="doi">10.1007/978-1-4939-0554-6_12</citation>
    </citations>

</tool>