<tool id="roary" name="Roary" version="3.13.0+galaxy0">
    <description>the pangenome pipeline - Quickly generate a core gene alignment from gff3 files</description>

    <requirements>
      <requirement type="package" version="3.13.0">roary</requirement>
    </requirements>

    <command detect_errors="exit_code"><![CDATA[
        #set $filenames = list()
        #for $gff in $gff_input.gffs
          cp '$gff' '${gff.element_identifier}.gff' &&
          #set $filename = str($gff.element_identifier) + '.gff'
          $filenames.append(str($filename))
        #end for

        roary
          -f out
          -p \${GALAXY_SLOTS:-1}
          -e
          -n
          -i '$percent_ident'
          -cd '$core_diff'
          -g '$advanced.maxclust'
          $advanced.split_para
          -t '$advanced.trans_tab'
          -iv '$advanced.mcl'

          #for $f in $filenames
            '$f'
          #end for

    ]]></command>

    <inputs>
      <conditional name="gff_input">
        <param name="gff_input_selector" type="select" label="Individual gff files or a dataset collection" help="Select between individual gff files or a collection of gff files">
          <option value="individual">Individual</option>
          <option value="collection">Collection</option>
        </param>
        <when value="individual">
          <param name="gffs" type="data" multiple="true" format="gff,gff3" label="select gff inputs to Roary" help="Select the files you wish to send to Roary, must be in gff3 format with the sequence data at the end of the file." />
        </when>
        <when value='collection'>
          <param name="gffs" type="data_collection" collection_type="list" format="gff,gff3" label="Dataset collection to submit to Roary" help="A dataset list collection of gff3 files to send to Roary for analysis." />
        </when>
      </conditional>
      <param name="percent_ident" type="integer" label="minimum percentage identity for blastp" help="Sets the minimum percentage identity for protein matches" value="95"/>
      <param name="core_diff" type="float" label="percentage of isolates a gene must be in to be core" help="The total percentage of the isolates that must have the gene for it to be considered a core gene." value="99.0"/>

      <param name="outputs" type="select" multiple="true" display="checkboxes" label="Additional outputs" help="Summary statistics, core gene alignments and Gene Presence/Absence tables are always produced.">
        <option value="abg_nw">Accessory binary genes in newick format</option>
        <option value="abg_fa">Accessory binary genes in fasta format</option>
        <option value="accgraph">Accessory graph in dot format</option>
        <option value="acchead_embl">Accessory header file in embl format</option>
        <option value="acctab">Accessory gene table in tabular format</option>
        <option value="blastfreq">Blast Identity Frequency Rtab file</option>
        <option value="clust">Clustered proteins file</option>
        <option value="coreaccgraph">Core Accessory Graph in dot format</option>
        <option value="coreaccembl">Core Accessory table in embl format</option>
        <option value="coreacctab">Core Accessory table in tabular format</option>
        <option value="genepa_rtab">Gene Presence Absence file in Rtab format</option>
        <option value="numcons_rtab">Number of Conserved Genes in Rtab format</option>
        <option value="numpangene_rtab">Number of Genes in Pan Genome in Rtab format</option>
        <option value="numnew_rtab">Number of New Genes in Rtab format</option>
        <option value="numuniq_rtab">Number of Unique Genes in Rtab format</option>
        <option value="pangenomeref">FASTA file which contains a single representative nucleotide sequence from each of the clusters in the pan genome (core and accessory)</option>
      </param>

      <section name="advanced" title="Advanced options" expanded="false">
        <param name="maxclust" type="integer" value="50000" label="Maximum number of clusters" help="The maximum number of clusters to assign proteins" />
        <param name="split_para" type="boolean" checked="False" truevalue="-s" falsevalue="" label="Don't split paralogs" help="Click yes so that paralogs aren't split."/>
        <param name="trans_tab" type="integer" value="11" label="Translation table to use [1 or 4 or 11]" help="DNA -> RNA translation table to use. Roary is designed for prokaryotes and so tables 1, 4 or 11 will work, others could be problematic." />
        <param name="mcl" type="float" value="1.5" label="MCL inflation value" help="This is an advanced setting, change only if you know what you are doing and then at your own risk!" />
      </section>

    </inputs>

    <outputs>
      <data format="tabular" name="sumstats" label="${tool.name} on ${on_string} Summary statistics" from_work_dir="out/summary_statistics.txt" />
      <data format="fasta" name="core_gene_aln" label="${tool.name} on ${on_string} Core Gene Alignment" from_work_dir="out/core_gene_alignment.aln" />
      <data format="csv" name="gene_p_a" label="${tool.name} on ${on_string} Gene Presence Absence" from_work_dir="out/gene_presence_absence.csv" />
      <data format="fasta" name="acc_bin" label="${tool.name} on ${on_string} Accessory Binary Genes" from_work_dir="out/accessory_binary_genes.fa" >
        <filter>outputs and 'abg_fa' in outputs</filter>
      </data>
      <data format="nhx" name="acc_bin_new" label="${tool.name} on ${on_string} Accessory Binary Genes (Newick)" from_work_dir="out/accessory_binary_genes.fa.newick" >
        <filter>outputs and 'abg_nw' in outputs</filter>
      </data>
      <data format="dot" name="acc_graph" label="${tool.name} on ${on_string} Acsessory Graph" from_work_dir="out/accessory_graph.dot">
        <filter>outputs and 'accgraph' in outputs</filter>
      </data>
      <data format="embl" name="acc_head_embl" label="${tool.name} on ${on_string} Accessory Header" from_work_dir="out/accessory.header.embl" >
        <filter>outputs and 'acchead_embl' in outputs</filter>
      </data>
      <data format="tabular" name="acc_tab" label="${tool.name} on ${on_string} Accessory Gene Table" from_work_dir="out/accessory.tab" >
        <filter>outputs and 'acctab' in outputs</filter>
      </data>
      <data format="txt" name="blast_freq" label="${tool.name} on ${on_string} Blast Identity Frequencies" from_work_dir="out/blast_identity_frequency.Rtab" >
        <filter>outputs and 'blastfreq' in outputs</filter>
      </data>
      <data format="txt" name="clust_file" label="${tool.name} on ${on_string} Clustered Proteins" from_work_dir="out/clustered_proteins" >
        <filter>outputs and 'clust' in outputs</filter>
      </data>
      <data format="dot" name="core_acc_graph" label="${tool.name} on ${on_string} Core Accessory Graph" from_work_dir="out/core_accessory_graph.dot" >
        <filter>outputs and 'coreaccgraph' in outputs</filter>
      </data>
      <data format="embl" name="core_acc_embl" label="${tool.name} on ${on_string} Core Accessory EMBL" from_work_dir="out/core_accessory.header.embl" >
        <filter>outputs and 'coreaccembl' in outputs</filter>
      </data>
      <data format="tabular" name="core_acc_tab" label="${tool.name} on ${on_string} Core Accessory Table" from_work_dir="out/core_accessory.tab" >
        <filter>outputs and 'coreacctab' in outputs</filter>
      </data>
      <data format="txt" name="gene_p_a_rtab" label="${tool.name} on ${on_string} Gene Presence Absence Rtab" from_work_dir="out/gene_presence_absence.Rtab" >
        <filter>outputs and 'genepa_rtab' in outputs</filter>
      </data>
      <data format="txt" name="num_cons_rtab" label="${tool.name} on ${on_string} Number of Conserved Genes" from_work_dir="out/number_of_conserved_genes.Rtab" >
        <filter>outputs and 'numcons_rtab' in outputs</filter>
      </data>
      <data format="txt" name="num_pangene_rtab" label="${tool.name} on ${on_string} Number of Genes in Pan Geneome" from_work_dir="out/number_of_genes_in_pan_genome.Rtab" >
        <filter>outputs and 'numpangene_rtab' in outputs</filter>
      </data>
      <data format="txt" name="num_new_rtab" label="${tool.name} on ${on_string} Number of New Genes" from_work_dir="out/number_of_new_genes.Rtab" >
        <filter>outputs and 'numnew_rtab' in outputs</filter>
      </data>
      <data format="txt" name="num_uniq_rtab" label="${tool.name} on ${on_string} Number of Unique Genes" from_work_dir="out/number_of_unique_genes.Rtab" >
        <filter>outputs and 'numuniq_rtab' in outputs</filter>
      </data>
      <data format="fasta" name="pan_genome_ref" label="${tool.name} on ${on_string} pan-genome reference" from_work_dir="out/pan_genome_reference.fa" >
        <filter>outputs and 'pangenomeref' in outputs</filter>
      </data>
    </outputs>

    <tests>
      <test>
        <param name="gff_input_selector" value="individual"/>
        <param name="gffs" value="ex1.gff,ex2.gff" ftype="gff3"/>
        <output name="sumstats" file="out/summary_statistics.txt" ftype="tabular"/>
      </test>
      <test>
        <param name="gff_input_selector" value="individual"/>
        <param name="gffs" value="ex1.gff,ex2.gff" ftype="gff3"/>
        <param name="percent_ident" value="50"/>
        <param name="core_diff" value="50.0"/>
        <output name="sumstats" file="test2/summary_statistics.txt" ftype="tabular"/>
      </test>
    </tests>

    <help><![CDATA[
**Roary**

Roary is a high speed stand alone pan genome pipeline, which takes annotated assemblies in GFF3 format (produced by Prokka) and calculates the pan genome. Using a standard desktop PC, it can analyse datasets with thousands of samples, something which is computationally infeasible with existing methods, without compromising the quality of the results. 128 samples can be analysed in under 1 hour using 1 GB of RAM and a single processor. To perform this analysis using existing methods would take weeks and hundreds of GB of RAM.

To use Roary, select two or more gff3 files OR a collection of gff3 files

**Options**:


       - Minimum percentage identity for blastp - an integer, default is [95]


       - Percentage of isolates a gene must be in to be core - a float, default is [99.0]


**Advanced Options**:

       - Maximum number of clusters - integer, default is [50000]


       - Don't split paralogs - check box


       - Translation table - which translation table to use, an integer, default is [11]


       - Change the MCL inflation value - a float, default is [1.5]


For further info see: http://sanger-pathogens.github.io/Roary/

    ]]></help>

    <citations>
        <citation type="doi">10.1093/bioinformatics/btv421</citation>
    </citations>

</tool>