<tool id="nlstradamus" name="NLStradamus" version="0.0.3">
    <description>Find nuclear localization signals (NLSs) in protein sequences</description>
    <command>
      NLStradamus -i $fasta_file -t $threshold -m $model -a $algorithm -tab > $tabular_file
    </command>
    <inputs>
        <param name="fasta_file" type="data" format="fasta" label="FASTA file of protein sequences"/> 
        <param name="model" type="select" display="radio" label="Model">
            <option value="1" selected="True">Two state</option>
            <option value="2">Four state</option>
        </param>
        <param name="algorithm" type="select" display="radio" label="Algorithm">
            <option value="0">Viterbi</option>
            <option value="1" selected="True">Posterior with threshold</option>
            <option value="2">Both</option>
        </param>
        <param name="threshold" type="float" label="Posterior theshold" value="0.6">
            <validator type="in_range" min="0" max="1" message="Threshold value should be between 0 and 1."/>
        </param>
    </inputs>
    <outputs>
        <data name="tabular_file" format="tabular" label="NLStradamus results" />
    </outputs>
    <requirements>
        <requirement type="binary">NLStradamus</requirement>
    </requirements>
    <tests>
    </tests>
    <help>
    
**What it does**

This calls the NLStradamus tool for prediction of nuclear localization
signals (NLSs), which uses a Hidden Markov Model (HMM).

The input is a FASTA file of protein sequences, and the output is tabular
with six columns (one row per NLS):

 * Sequence identifier
 * Algorithm (posterior or Viterbi)
 * Score (probability between threshold and 1 for posterior algorithm)
 * Start
 * End
 * Sequence of NLS

-----

**References**

A. N. Nguyen Ba, A. Pogoutse, N. Provart, A. M. Moses.
NLStradamus: a simple Hidden Markov Model for nuclear localization signal prediction.
BMC Bioinformatics. 2009 Jun 29;10(1):202.

http://www.moseslab.csb.utoronto.ca/NLStradamus

    </help>
</tool>