<tool id="bg_sortmerna" name="Filter with SortMeRNA" version="1.9.0">
    <description>Fast and accurate filtering of ribosomal RNAs in metatranscriptomic data</description>
    <requirements>
        <requirement type='package' version="1.9">sortmerna</requirement>
    </requirements>
    <stdio>
        <regex match="This program builds a Burst trie on an input rRNA database"
            source="both"
            level="fatal"
            description="Buildtrie program failed to execute." />
        <regex match="The database name"
            source="both"
            level="fatal"
            description="The database ${databases} has not been preprocessed using buildtrie before using SortMeRNA." />
    </stdio>
    <version_command>
<![CDATA[
sortmerna --version 2>&1|grep 'SortMeRNA version'
]]>
    </version_command>
    <command interpreter="python">
<![CDATA[
    sortmerna.py
    --sortmerna "
        $strand_search
        #if str( $read_family.read_family_selector ) == 'other':
            --I $input_reads -r $read_family.ratio_parameter
        #else:
            $read_family.read_family_selector $input_reads
        #end if

        #if str( $sequencing_type.sequencing_type_selector ) == 'paired':
            $sequencing_type.paired_type
        #end if

        #if $outputs_selected:
            #if 'accept' in $outputs_selected.value:
                --accept accept_file
            #end if
            #if 'other' in $outputs_selected.value:
                --other other_file
            #end if
        #end if

        $log
        -a \${GALAXY_SLOTS:-4}
        "
        #if str( $databases_type.databases_selector ) == 'history':
            --buildtrie
            #for $db in $databases_type.input_databases
                $db.database_name
            #end for
        #else:
            ## databases path is not directly accessible, must match by hand with LOC file contents
            ${' '.join([dict([(x[0], x[2]) for x in $databases_type.input_databases.input.options.tool_data_table.data])[y]
                       for y in $databases_type.input_databases.value])}
        #end if
]]>
    </command>
    <inputs>
    <conditional name="read_family">
        <param name="read_family_selector" type="select" format="text" label="Sequencing technology of querying sequences (reads)"
            help="The Illumina platform is more common for large scale metatranscriptomic projects requiring a high throughput.">
            <option value="--I">Illumina Solexa</option>
            <option value="--454">454 Roche</option>
            <option value="other">Other</option>
        </param>
        <when value="other">
            <param name="ratio_parameter" type="float" value="1" min="0" max="1"
                label="Ratio parameter (the number of hits on the read / read length)"
                help="The ratio parameter for SortMeRNA has been set to r=0.25 for Illumina Solexa reads and to r=0.15 for 454 Roche reads.
                    For other read types, if the sequencing technology produces high quality reads with a low substitution error rate
                    (0.1 substitutions per 100 bases, such as Illumina), then the ratio parameter can be set to r=[0.23,0.27].
                    If the sequencing technology has a high indel error rate (1-2 indels per 100 bases, such as 454 or Ion Torrent),
                    then the ratio parameter can be set to r=[0.13,0.17] (-r)."/>
        </when>
    </conditional>
    <param format="fasta,fastq" name="input_reads" type="data" label="Querying sequences (reads)" help=""/>
    <conditional name="sequencing_type">
        <param name="sequencing_type_selector" type="select" label="Sequencing type">
            <option value="not_paired">Reads are not paired</option>
            <option value="paired">Reads are paired</option>
        </param>
        <when value="paired">
            <param name="paired_type" type="select" display="radio" label="If one read of a pair is accepted and the other not, output both reads"
                help="SortMeRNA does not use the pairing information for filtering RNA,
                    however if one read of a pair is accepted and the other is not,
                    the resulting output may break apart the pair into two separate files.
                    The purpose of 'Reads are paired' option is to preserve the pairing of the reads.">
                <option value="--paired-in">to accepted file (--paired-in)</option>
                <option value="--paired-out">to rejected file (--paired-out)</option>
            </param>
        </when>
    </conditional>

    <param name="strand_search" type="select" label="Which strands to search" display="radio">
        <option value="">Search both strands</option>
        <option value="-F">Search only the forward strand (-F)</option>
        <option value="-R">Search only the reverse-complementary strand (-R)</option>
    </param>

    <conditional name="databases_type">
        <param name="databases_selector" type="select" label="Databases to query"
            help="Public rRNA databases provided with SortMeRNA have been indexed.
                On the contrary, personal databases must be indexed each time SortMeRNA is launched.
                Please be patient, this may take some time depending on the size of the given database.">
            <option value="cached" selected="true">Public ribosomal databases</option>
            <option value="history">Databases from your history</option>
        </param>
        <when value="cached">
            <param name="input_databases" label="rRNA database" type="select" display="checkboxes" multiple="true">
                <options from_data_table="rRNA_databases" />
                <validator type="no_options" message="Select at least one database"/>
            </param>
        </when>
        <when value="history">
            <repeat name="input_databases" title="Database" min="1">
                <param name="database_name" type="data" format="fasta" label="rRNA database"
                    help="Your database will be indexed first, which may take up to several minutes."/>
            </repeat>
        </when>
    </conditional>

    <!-- Outputs -->
    <param name="outputs_selected" type="select" display="checkboxes" multiple="true" label="Output options">
        <option value="accept" selected="True">Reads matching to at least one database</option>
        <option value="other">Reads not found in any database</option>
    </param>
    <param name="log" type="boolean" checked="False" truevalue="--log log_file" falsevalue="" label="Statistics file"
           help="Generates statistics for the rRNA content of reads, as well as rRNA subunit distribution. (--log)">
    </param>

    </inputs>
    <outputs>
        <data format_source="input_reads" name="output_accept" from_work_dir="accept_file.dat"
            label="Matching reads on ${on_string} (${input_reads.datatype.file_ext})">
            <filter>outputs_selected and 'accept' in outputs_selected</filter>
        </data>
        <data format_source="input_reads" name="output_other" from_work_dir="other_file.dat"
            label="Reads not found on ${on_string} (${input_reads.datatype.file_ext})">
            <filter>outputs_selected and 'other' in outputs_selected</filter>
        </data>
        <data format="txt" name="output_log" label="${tool.name} statistics (txt)" from_work_dir="log_file.log">
            <filter>log</filter>
        </data>
    </outputs>
    <tests>
        <test>
            <param name="read_family_selector" value="I" />
            <param name="input_reads" value="sortmerna_wrapper_in1.fastq" />
            <param name="sequencing_type_selector" value="not_paired" />
            <param name="strand_search" value="" />
            <param name="databases_selector" value="cached" />
            <param name="input_databases" value="rfam-5.8s,rfam-5s" />
            <param name="outputs_selected" value="accept,other" />
            <param name="log" value="" />
            <param name="options_type_selector" value="less" />
            <output name="output_accept" file="sortmerna_wrapper_accept1.fastq" />
            <output name="output_other" file="sortmerna_wrapper_other1.fastq" />
        </test>
    </tests>
    <help>
<![CDATA[
**What it does**

SortMeRNA_ is a software designed to rapidly filter ribosomal RNA fragments
from metatransriptomic data produced by next-generation sequencers.
It is capable of handling large RNA databases and sorting out all fragments
matching to the database with high accuracy and specificity.

.. _SortMeRNA: http://bioinfo.lifl.fr/RNA/sortmerna/


**Input**

The input is one file of reads in FASTA or FASTQ format and any number of rRNA databases to search against.
If the user has two foward-reverse paired-sequencing reads files, they may use
the script "merge_paired_reads.sh" to interleave the reads into one file, preserving their order.

If the sequencing type for the reads is paired-ended, the user has two options under
"Sequencing type" to filter the reads and preserve their order in the file.
For a further example of each option, please refer to Section 4.2.3 in the `SortMeRNA User Manual`_.

.. _sortmerna user manual: http://bioinfo.lifl.fr/RNA/sortmerna/code/SortMeRNA-user-manual-v1.7.pdf


**Output**

The output will follow the same format (FASTA or FASTQ) as the reads. Optionally, a statistic file for the rRNA content of reads, as well as rRNA subunit distribution can be generated.


**rRNA databases**

SortMeRNA is distributed with 8 representative rRNA databases, which were
all constructed from the SILVA SSU,LSU (version 111) and the RFAM 5/5.8S
(version 11.0) databases using the tool UCLUST.

+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| Representative database  | id % | average id% | # seq (clustered) | Origin                 |  # seq (original) |
+==========================+======+=============+===================+========================+===================+
| SILVA 16S bacteria       |   85 |        91.6 |              8174 | SILVA SSU Ref NR v.111 |            244077 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| SILVA 16S archaea        |   95 |        96.7 |              3845 | SILVA SSU Ref NR v.111 |             10919 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| SILVA 18S eukarya        |   95 |        96.7 |              4512 | SILVA SSU Ref NR v.111 |             31862 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| SILVA 23S bacteria       |   98 |        99.4 |              3055 | SILVA LSU Ref v.111    |             19580 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| SILVA 23s archaea        |   98 |        99.5 |               164 | SILVA LSU Ref v.111    |               405 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| SILVA 28S eukarya        |   98 |        99.1 |              4578 | SILVA LSU Ref v.111    |              9321 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| Rfam 5S archaea/bacteria |   98 |        99.2 |             59513 | RFAM                   |            116760 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+
| Rfam 5.8S eukarya        |   98 |        98.9 |             13034 | RFAM                   |            225185 |
+--------------------------+------+-------------+-------------------+------------------------+-------------------+

id %: members of the cluster must have identity at least 'id %' identity with the representative sequence

average id %: average identity of a cluster member to the representative sequence

The user may also choose to use their own rRNA databases.

.. class:: warningmark

Note that your personal databases are indexed each time, and that
this may take some time depending on the size of the given database.
]]>
    </help>

    <citations>
        <citation type="doi">10.1093/bioinformatics/bts611</citation>
        <citation type="doi">10.1093/nar/gks1219</citation>
        <citation type="doi">10.1093/nar/gks1005</citation>
        <citation type="doi">10.1093/bioinformatics/btq461</citation>
        <citation type="doi">10.1038/nbt.2198</citation>
    </citations>
</tool>