<?xml version="1.0" encoding="utf-8"?>
<tool id="sortmerna_wrapper" version="1.0" name="Filter with SortMeRNA">
  <requirements>
    <requirement type='package' version="1.7">sortmerna</requirement>
  </requirements>
  <description>Fast and accurate filtering of ribosomal RNAs in metatranscriptomic data</description>
  <command interpreter="python">
sortmerna_wrapper.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
#if str( $options.options_type_selector ) == 'more':
	-a $options.number_of_threads
#end if
"
#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"
	   help="The Illumina platform is more common for large scale metatranscriptomic projects requiring a high throughput.">
	<label>Sequencing technology of querying sequences (reads)</label>
	<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]."/>
      </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" label="If one read of a pair is accepted and the other not, output both reads" display="radio"
	       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</option>
	  <option value="--paired-out">to rejected file</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</option>
      <option value="-R">Search only the reverse-complementary strand</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.">
    </param>

    <!-- Advanced options -->
    <conditional name="options">
      <param name="options_type_selector" type="select" label="Advanced Options">
	<option value="less" selected="True">Less options</option>
	<option value="more">More options</option>
      </param>
      <when value="less">
	<!-- no options -->
      </when>
      <when value="more">
	<param name="number_of_threads" type="integer" label="Number of threads to use" value="1" min="1"/>
      </when>
    </conditional>
  </inputs>
  <outputs>
    <data format="input" 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="input" 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>
  <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>
  <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>
**Overview**

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/

If you use this tool, please cite Kopylova E., Noé L. and Touzet H.,
`"SortMeRNA: Fast and accurate filtering of ribosomal RNAs in metatranscriptomic data"`__,
Bioinformatics (2012), doi: 10.1093/bioinformatics/bts611.

.. __: http://bioinformatics.oxfordjournals.org/content/28/24/3211

------

**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.

In the standalone version of SortMeRNA, the user may output the matching reads in a separate file per database (--bydbs option). This option will be made available in a future version of Galaxy.

------

**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 % | avergage id% | # seq | Origin                 |  # seq | filtered to remove |
+==========================+======+==============+=======+========================+========+====================+
| SILVA 16S bacteria       |   85 |         91.6 |  8174 | SILVA SSU Ref NR v.111 | 244077 | 23s                |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| SILVA 16S archaea        |   95 |         96.7 |  3845 | SILVA SSU Ref NR v.111 |  10919 | 23s                |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| SILVA 18S eukarya        |   95 |         96.7 |  4512 | SILVA SSU Ref NR v.111 |  31862 | 26s,28s,23s        |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
|                                                                                                               |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| SILVA 23S bacteria       |   98 |         99.4 |  3055 | SILVA LSU Ref v.111    |  19580 | 16s,26s,28s        |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| SILVA 23s archaea        |   98 |         99.5 |   164 | SILVA LSU Ref v.111    |    405 | 16s,26s,28s        |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| SILVA 28S eukarya        |   98 |         99.1 |  4578 | SILVA LSU Ref v.111    |   9321 | 18s                |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
|                                                                                                               |
+--------------------------+------+--------------+-------+------------------------+--------+--------------------+
| 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.

------

**SortMeRNA parameter list**

The standalone, command-line version of SortMeRNA uses the following parameters.

For indexing (buildtrie):

This program builds a Burst trie on an input rRNA database file in fasta format
and stores the material in binary files under the folder '/automata'::

    ./buildtrie --db [path to rrnas database file name {.fasta}]  {OPTIONS}

The list of OPTIONS can be left blank, the default values will be used::

    -L  length of the sliding window (the seed)
        (default: 18)

    -F  search only the forward strand
    -R  search only the reverse-complementary strand
        (default: both strands are searched)

    -h  help




For sorting (sortmerna):

To run SortMeRNA, type in any order after 'sortmerna'::

        --I      [illumina reads file name {fasta/fastq}]

        --454    [roche 454 reads file name {fasta/fastq}]

        -n       number of databases to use (must precede --db)

        --db     [rrnas database name(s)]

                 One database,
                 ex 1. -n 1 --db /path1/database1.fasta

                 Multiple databases,
                 ex 2. -n 2 --db /path2/database2.fasta /path3/database3.fasta

        {OPTIONS}

The list of OPTIONS can be left blank, the default values will be used::

        --accept      [accepted reads file name]
        --other       [rejected reads file name]
                      (default: no output file is created)

        --bydbs       output the accepted reads by database
                      (default: concatenated file of reads)

        --log         [overall statistics file name]
                      (default: no statistics file created)

        --paired-in   put both paired-end reads into --accept file
        --paired-out  put both paired-end reads into --other file
                      (default: if one read is accepted and the other is not,
                      separate the reads into --accept and --other files)

        -r            ratio of the number of hits on the read / read length
                      (default Illumina: 0.25, Roche 454: 0.15)

        -F            search only the forward strand
        -R            search only the reverse-complementary strand
                      (default: both strands are searched)

        -a            number of threads to use
                      (default: 1)

        -m            (m x 4096 bytes) for loading the reads into memory
                      ex. '-m 4' means 4*4096 = 16384 bytes will be allocated for the reads
                      note: maximum -m is 1020039
                      (default: m = 262144 = 1GB)

        -v            verbose
                      (default: deactivated)

        -h            help

        --version     version number

------

**Bibliography**

[1] Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO (2013) The SILVA ribosomal RNA gene database project: improved data processing and web-based tools, Nucleic Acids Research, 41 (D1): D590-D596.

[2] Rfam 11.0: 10 years of RNA families. S.W. Burge, J. Daub, R. Eberhardt, J. Tate, L. Barquist, E.P. Nawrocki, S.R. Eddy, P.P. Gardner, A. Bateman. Nucleic Acids Research (2012),  doi: 10.1093/nar/gks1005

[3] Edgar, R.C. (2010) Search and clustering orders of magnitude faster than BLAST, Bioinformatics 26(19), 2460-2461, doi: 10.1093/bioinformatics/btq461

[4] Loman, N. J. and Misra, Raju V and Dallman, Timothy J and Constantinidou, Chrystala and Gharbia, Saheer E and Wain, John and Pallen, Mark J., Performance comparison of benchtop high-throughput sequencing platforms (2012), Nature Biotechnology, 30 (5). pp. 434-439
  </help>
</tool>