<tool id="metaphlan2" name="MetaPhlAn2" version="2.5.0">

    <description>to profile the composition of microbial communities</description>

    <macros>
        <import>metaphlan2_macros.xml</import>
    </macros>

    <expand macro="requirements"/>
    <expand macro="stdio"/>

    <version_command>
<![CDATA[
metaphlan2.py -v
]]>
    </version_command>

    <command>
<![CDATA[
        (which bowtie2 || exit 200)

        &&

        #if $db.db_selector == "history"
            mkdir ref_db
            &&
            bowtie2-build $db.db_sequences ref_db/ref_db
            &&
            python $__tool_directory__/transform_json_to_pkl.py
                --json_input $db_metadata
                --pkl_output ref_db/metadata.pkl
            &&
        #end if

        metaphlan2.py
            $input_file
            -o $output_file
            --input_type ${input_file.datatype.file_ext}

            --bowtie2_exe `which bowtie2`

            #if $db.db_selector == "cached"
                #set $table = dict([(_[0], _[2]) for _ in $db.cached_db.input.options.tool_data_table.data])
                #set $db_choice = $db.cached_db.value
                --bowtie2db $table[$db_choice]
                --mpa_pkl $table[$db_choice]".pkl"
            #else
                --bowtie2db ref_db/ref_db
                --mpa_pkl ref_db/metadata.pkl
            #end if

            --no_map

            -t $analysis_type.analysis_type_select
            #if $analysis_type.analysis_type_select == "rel_ab"
                --tax_lev $analysis_type.taxonomic_level
            #else if $analysis_type.analysis_type_select == "marker_ab_table"
                --nreads $analysis_type.nreads
            #else if $analysis_type.analysis_type_select == "marker_pres_table"
                --pres_th $analysis_type.pres_th
            #end if

            --min_cu_len $min_cu_len
            --min_alignment_len $min_alignment_len

            $ignore_viruses
            $ignore_eukaryotes
            $ignore_bacteria
            $ignore_archaea

            --stat_q $stat_q
            -s $sam_output_file
]]>
    </command>

    <inputs>
        <param name="input_file" type="data" format="fastq,fasta,sam" label="Input file" help=""/>

        <conditional name="db">
            <param name="db_selector" type="select" label="Database with clade-specific marker genes" help="">
                <option value="cached" selected="true">Locally cached</option>
                <option value="history">From history</option>
            </param>

            <when value="cached">
                <param name="cached_db" label="Cached database with clade-specific marker genes" type="select" >
                <options from_data_table="metaphlan2_db" />
                </param>
            </when>
            <when value="history">
                <param name="db_sequences" type="data" format="fasta" label="Database with clade-specific marker genes from history" help="(--bowtie2db)"/>
                <param name="db_metadata" type="data" format="json" label="Metadata associate to the database with clade-specific marker genes from history" help="(--mpa_pkl)"/>
            </when>
        </conditional>

        <conditional name="analysis_type">
            <param name="analysis_type_select" type="select" label="Type of analysis to perform" help="(-t)">
              <option value="rel_ab" selected="true">Profiling a metagenomes in terms of relative abundances</option>
              <option value="reads_map">Mapping from reads to clades (only reads hitting a marker)</option>
              <option value="clade_profiles">Normalized marker counts for clades with at least a non-null marker</option>
              <option value="marker_ab_table">Normalized marker counts (only when > 0.0 and normalized by metagenome size if --nreads is specified)</option>
              <option value="marker_counts">Non-normalized marker counts (use with extreme caution)</option>
              <option value="marker_pres_table">List of markers present in the sample (threshold at 1.0 if not differently specified with --pres_th</option>
            </param>

            <when value="rel_ab">
              <param name="taxonomic_level" type="select" label="Taxonomic level for the relative abundance output" help="(--tax_lev)">
                <option value="a" selected="true">All taxonomic levels</option>
                <option value="k">Kingdoms (Bacteria and Archaea) only</option>
                <option value="p">Phyla only</option>
                <option value="c">Classes only</option>
                <option value="o">Orders only</option>
                <option value="f">Families only</option>
                <option value="g">Genera only</option>
                <option value="s">Species only</option>
              </param>
            </when>

            <when value="reads_map"/>
            <when value="clade_profiles"/>

            <when value="marker_ab_table">
                <param name="nreads" type="integer" value="0" label="Total number of reads in the original metagenome" help="It is used for normalizing the length-normalized counts with the metagenome size as well. No normalization applied if the value is not specified"/>
            </when>

            <when value="marker_counts"/>

            <when value="marker_pres_table">
                <param name="pres_th" type="integer" value="0" label=" Threshold for calling a marker present" help=""/>
            </when>
        </conditional>

        <param name="min_cu_len" type="integer" value="2000" label="Minimum total nucleotide length for the markers in a clade for estimating the abundance without considering sub-clade abundances" help=""/>

        <param name="min_alignment_len" type="integer" value="0" label="Sam records for aligned reads with the longest subalignment length smaller than this threshold will be discarded." help=""/>

        <param name="ignore_viruses" type='boolean' checked="true" truevalue='' falsevalue='--ignore_viruses' label="Profile viral organisms?" help="" />
        <param name="ignore_eukaryotes" type='boolean' checked="true" truevalue='' falsevalue='--ignore_eukaryotes' label="Profile eukaryotic organisms?" help="" />

        <param name="ignore_bacteria" type='boolean' checked="true" truevalue='' falsevalue='--ignore_bacteria' label="Profile bacteria organisms?" help="" />

        <param name="ignore_archaea" type='boolean' checked="true" truevalue='' falsevalue='--ignore_archaea' label="Profile archea organisms?" help="" />

        <param name="stat_q" type="float" value="0.1" label="Quantile value for the robust average" help=""/>
    </inputs>

    <outputs>
        <data format="tabular" name="output_file" label="${tool.name} on ${on_string}: Community profile" />
        <data format="sam" name="sam_output_file" label="${tool.name} on ${on_string}: Sam file" />
    </outputs>

    <tests>
        <test>
            <param name="input_file" value="input_sequences.fasta"/>
            <param name="db_selector" value="history" />
            <param name="db_metadata" value="marker_metadata.json" />
            <param name="db_sequences" value="marker_sequences.fasta" />
            <param name="analysis_type_select" value="rel_ab" />
            <param name="taxonomic_level" value="a" />
            <param name="min_cu_len" value="2000" />
            <param name="min_alignment_len" value="0" />
            <param name="ignore_viruses" value="" />
            <param name="ignore_eukaryotes" value="" />
            <param name="ignore_bacteria" value="" />
            <param name="ignore_archaea" value="" />
            <param name="stat_q" value="0.1" />
            <output name="output_file" file="community_profile.tabular"/>
        </test>
    </tests>

    <help><![CDATA[
**What it does**

MetaPhlAn is a computational tool for profiling the composition of microbial communities (Bacteria, Archaea, Eukaryotes and Viruses) from metagenomic shotgun sequencing data with species level resolution. For more information, check the `user manual <https://bitbucket.org/biobakery/metaphlan2/>`_.

**Inputs**

Metaphlan2 takes as input a sequence file in fasta, fastq, a BowTie2 produced SAM file.

It is also possible to use a custom database with clade-specific marker genes. In this case, a fasta file with marker gene sequences is required and also a file containing metadata. This file is a json file with:

::

  {
    "taxonomy": {
            "taxonomy of genome1": genome1_length,
            "taxonomy of genome2": genome2_length,
            ...
        }
    "markers": {
            "marker1_name": {
                "clade": the clade that the marker belongs to,
                "ext": [list of external genomes where the marker appears],
                "len": length of the marker,
                "score": score of the marker,
                "taxon": the taxon of the marker
            }
            ...
        }
  }

The marker names correspond to sequence name in corresponding fasta file with marker gene sequences.

**Outputs**

The main output file is a tab-separated output file of the predicted taxon relative abundances.

    ]]></help>

    <citations>
        <citation type="doi">10.1038/nmeth.3589</citation>
    </citations>
</tool>