<tool id="motiffinding_weeder2" name="Weeder2" version="2.0.3">
  <description>Motif discovery in sequences from coregulated genes of a single species</description>
  <requirements>
    <requirement type="package" version="2.0">weeder</requirement>
  </requirements>
  <command detect_errors="exit_code"><![CDATA[
  bash $__tool_directory__/weeder2_wrapper.sh
  $sequence_file $species_code ${species_code.fields.path}
  $output_motifs_file $output_matrix_file
  $strands
  #if $chipseq.use_chipseq
     -chipseq -top $chipseq.top
  #end if
  #if str( $advanced_options.advanced_options_selector ) == "on"
     -maxm $advanced_options.n_motifs_report
     -b $advanced_options.n_motifs_build
     -sim $advanced_options.sim_threshold
     -em $advanced_options.em_cycles
  #end if
  ]]></command>
  <inputs>
    <param name="sequence_file" type="data" format="fasta" label="Input sequence" />
    <param name="species_code" type="select" label="Species to use for background comparison">
      <options from_data_table="weeder2">
      </options>
    </param>
    <param name="strands" label="Use both strands of sequence" type="boolean"
	   truevalue="" falsevalue="-ss" checked="True"
	   help="If not checked then use -ss option" />
    <conditional name="chipseq">
      <param name="use_chipseq" type="boolean"
	     label="Use the ChIP-seq heuristic"
	     help="Speeds up the computation (-chipseq)"
	     truevalue="yes" falsevalue="no" checked="on" />
      <when value="yes">
	<param name="top" type="integer" value="100"
	       label="Number of top input sequences with oligos to scan for"
	       help="Increase this value to improve the chance of finding motifs enriched only in a subset of your input sequences (-top)" />
      </when>
      <when value="no"></when>
    </conditional>
    <conditional name="advanced_options">
      <param name="advanced_options_selector" type="select"
	     label="Display advanced options">
	<option value="off">Hide</option>
	<option value="on">Display</option>
      </param>
      <when value="on">
	<param name="n_motifs_report" type="integer" value="25"
	       label="Number of discovered motifs to report" help="(-maxm)" />
	<param name="n_motifs_build" type="integer" value="50"
	       label="Number of top scoring motifs to build occurrences matrix profiles and outputs for"
	       help="(-b)" />
	<param name="sim_threshold" type="float" min="0.0" max="1.0" value="0.95"
	       label="Similarity threshold for the redundancy filter"
	       help="Remove motifs that are too similar, with lower values imposing a stricter filter. Must be between 0.0 and 1.0 (-sim)" />
	<param name="em_cycles" type="integer" min="0" max="100" value="1"
	       label="Number of expectation maximization (EM) cycles to perform"
	       help="Number of cycles must be between 0 and 100 (-em)" />
      </when>
      <when value="off">
      </when>
    </conditional>
  </inputs>
  <outputs>
    <data name="output_motifs_file" format="txt" label="Weeder2 on ${on_string} (motifs)" />
    <data name="output_matrix_file" format="txt" label="Weeder2 on ${on_string} (matrix)" />
  </outputs>
  <tests>
    <test>
      <param name="sequence_file" value="weeder_in.fa" ftype="fasta" />
      <param name="species_code" value="MM" />
      <output name="output_motifs_file" file="weeder2_motifs.out" lines_diff="2" />
      <output name="output_matrix_file" file="weeder2_matrix.out" />
    </test>
  </tests>
  <help>

.. class:: infomark

**What it does**

Weeder2 is a program for finding novel motifs (transcription factor binding sites)
conserved in a set of regulatory regions of related genes.

-------------

.. class:: infomark

**Usage advice**

Guidelines on how to use this tool can be seen in Zambelli et al. 2014 (see link
below), but the following is a brief guide. Please note that **motifs** are a model
or matrix that describes a set of sequences that may differ in the base composition.
**Oligos** are specific sequences found within the input sequences or genomic
background.

**Input sequence** (in FASTA format) should be short (100-200bp) and be reasonably
expected to contain an enriched motif(s).  This is not generally an issue with
transcription factor ChIP-seq derived sequences centred on the summit of binding
regions that are expected to contain a dominant motif and possibly secondary motifs.

There is **no need to mask sequence for repetitive sequence** as factors may
legitimately bind repetitive sequence.

**Use both strands of sequence** by default, unless there is a specific reason not
to do so.

**Species to use for background comparison** should match the genome used to
generate the **input sequence**. The background genome motif frequencies are
generated from within the promoter regions of annotated genes and are shown to be a
good background for both promoter and other regulatory regions. 

**Use the ChIP-seq heuristic** (-chipseq) when there are a large number of
input sequences (hundreds or thousands). When -chipseq is used Weeder will use
only oligos from the first 100 sequences to build motifs with which it scans
all of the input sequences. This speeds up the computational time without too much
risk of losing important motifs. Even if not strictly necessary it's advisable to
order input sequences by their significance, e.g. fold enrichment or Pvalue. For
large data sets (-top) should be set to a number equating at least 10 to 20% of
input sequences (as recommended by the authors).

**Number of discovered motifs to report** (-maxm) limits the number of reported
motifs even if there are more than -maxm. **Number of top scoring motifs to build
occurrences matrix profiles and outputs for** (-b) changes the number of top
scoring motifs of length 6, 8 and 10 for which the occurrence matrix is built.
Increasing -b may result in a larger number of reported motifs, but with potentially
more of low significance and increases the computational time. If increasing -b does
not result in more motifs in your results it means that the additional motifs are
filtered out by the redundancy filter or that the maximum number of reported motifs
set by -maxm has been reached.

**Similarity threshold for the redundancy filter** (-sim) default setting is
recommended.

**Number of expectation maximization (EM) cycles to perform** (-em) default is
recommended.  The option is included to help "clean up" the resulting motif matrices.
In this version the number of EM steps can be increased, which can be useful for
motifs with highly redundant stretches of sequence.

-------------

.. class:: infomark

**A note on the results**

The resulting matrices are the result of scanning (by default both strands) for
oligos of length 6, 8 and 8, allowing 1, 2 and 3 substitutions respectively. The
matrices within the matrix.w2 file can be input into other tools. The recommended
next step is to use **STAMP** (http://www.benoslab.pitt.edu/stamp/), which displays
the motifs as logos and identifies matches with libraries of known DNA binding
motifs, such as TRANSFAC or JASPAR.

-------------

.. class:: infomark

**Credits**

This Galaxy tool has been developed by Peter Briggs and Ian Donaldson within the
Bioinformatics Core Facility at the University of Manchester, and runs the Weeder2
motif discovery package:

 * Zambelli, F., Pesole, G. and Pavesi, G. 2014. Using Weeder, Pscan, and PscanChIP
   for the Discovery of Enriched Transcription Factor Binding Site Motifs in
   Nucleotide Sequences. Current Protocols in Bioinformatics. 47:2.11:2.11.1–2.11.31.
 * http://onlinelibrary.wiley.com/doi/10.1002/0471250953.bi0211s47/full

This tool is compatible with Weeder 2.0:

 * http://159.149.160.51/modtools/downloads/weeder2.html

Please kindly acknowledge both this Galaxy tool, the Weeder package and the utility
scripts if you use it in your work.
  </help>
  <citations>
    <!--
    See https://wiki.galaxyproject.org/Admin/Tools/ToolConfigSyntax#A.3Ccitations.3E_tag_set
    Can be either DOI or Bibtex
    Use http://www.bioinformatics.org/texmed/ to convert PubMed to Bibtex
    -->
    <citation type="doi">10.1002/0471250953.bi0211s47</citation>
  </citations>
</tool>