<?xml version="1.0"?>
<tool id="rost_reprof" name="reprof" version="@WRAPPER_VERSION@-gx1">
  <description>protein secondary structure and accessibility prediction</description>
  <macros>
    <import>macros.xml</import>
  </macros>
  <expand macro="requirements"/>
  <command detect_errors="aggressive"><![CDATA[
      ln -s $query query.fa;
      python $__tool_directory__/reprof.py query.fa \$REPROF_MODEL_DIR
]]></command>
  <inputs>
      <param name="query" label="Protein sequences" type="data" format="fasta" />
      <param name="outputs" type="select" multiple="true" label="Additional Outputs">
        <option value="pE">Probability of Extended/Sheet</option>
        <option value="pH">Probability of Helix</option>
        <option value="pL">Probability of Loop</option>
        <option value="RI_S">Secondary Structure Reliability Index</option>

        <option value="PACC">Absolute Solvent Accessibility</option>
        <option value="PREL">Relative Solvent Accessibility</option>
        <option value="P10">Relative Solvent Accessibility</option>
        <option value="RI_A">Solvent Accessibility Reliability Index</option>
      </param>
  </inputs>
  <outputs>
      <data format="gff3" name="secondary_structure" from_work_dir="secondary_structure.gff3" label="Secondary Structure of ${on_string}"/>
      <data format="wig" name="pE" from_work_dir="pE.wig" label="Probability of Extended/Sheet (${on_string})">
        <filter>outputs is not None and 'pE' in outputs</filter>
      </data>
      <data format="wig" name="pH" from_work_dir="pH.wig" label="Probability of Helix (${on_string})">
        <filter>outputs is not None and 'pH' in outputs</filter>
      </data>
      <data format="wig" name="pL" from_work_dir="pL.wig" label="Probability of Loop (${on_string})" >
        <filter>outputs is not None and 'pL' in outputs</filter>
      </data>
      <data format="wig" name="RI_S" from_work_dir="RI_S.wig" label="Secondary Structure Reliability Index (${on_string})">
        <filter>outputs is not None and 'RI_S' in outputs</filter>
      </data>

      <data format="gff3" name="solvent_accessibility" from_work_dir="solvent_accessibility.gff3" label="Solvent Accessibile Regions of ${on_string}" />
      <data format="wig" name="PACC" from_work_dir="PACC.wig" label="Absolute Solvent Accessibility of ${on_string}">
        <filter>outputs is not None and 'PACC' in outputs</filter>
      </data>
      <data format="wig" name="PREL" from_work_dir="PREL.wig" label="Relative Solvent Accessibility of ${on_string}">
        <filter>outputs is not None and 'PREL' in outputs</filter>
      </data>
      <data format="wig" name="P10" from_work_dir="P10.wig" label="Relative Solvent Accessibility (0-9) of ${on_string}">
        <filter>outputs is not None and 'P10' in outputs</filter>
      </data>
      <data format="wig" name="RI_A" from_work_dir="RI_A.wig" label="Solvent Accessibility Reliability Index (${on_string})">
        <filter>outputs is not None and 'RI_A' in outputs</filter>
      </data>

      <data format="tabular" name="report" from_work_dir="report.tsv" label="Report on Protein Composition of ${on_string}" />
  </outputs>
  <tests>
      <test>
          <param name="query" value="test.fa" />

          <output name="secondary_structure" file="secondary_structure.gff3" />
          <output name="solvent_accessibility" file="solvent_accessibility.gff3" />
          <output name="pE" file="pE.wig" />
          <output name="pH" file="pH.wig" />
          <output name="pL" file="pL.wig" />
          <output name="RI_S" file="RI_S.wig" />
          <output name="PACC" file="PACC.wig" />
          <output name="PREL" file="PREL.wig" />
          <output name="P10" file="P10.wig" />
          <output name="RI_A" file="RI_A.wig" />
      </test>
  </tests>
  <help><![CDATA[
reprof
======

Reprof is a protein secondary structure and accessibility predictor from the
Rost Lab. Prediction is done from protein sequences.

Three states of secondary structure are predicted: helix (H; includes alpha-,
pi- and 3_10-helix), (beta-)strand (E = extended strand in beta-sheet
conformation of at least two residues length) and loop (L).

Protein Classification
----------------------

Proteins can be classified as mixed given the following classes:

-  'all-alpha': %H > 45% AND %E < 5%
-  'all-beta': %H < 5% AND %E > 45%
-  'alpha-beta': %H > 30% AND %E > 20%
-  'mixed': All others

Gff3 Data and Evidence Codes
----------------------------

In order to facilitate interoperability with other downstream analyses, and to
document provenance, the Galaxy tool author has opted to apply ECO:0000255 to
all annotated features in the GFF3 file. Per the `UniPro Evidence page <http://www.uniprot.org/help/evidences#ECO:0000255>`__

    The ECO code ECO:0000255 is also used for information which has been
    generated by various sequence analysis programs that are used during the
    manual curation process and which has been verified by a curator.

      ]]></help>
  <citations>
    <citation type="doi">10.1186/1471-2105-10-S13-O3</citation>
    <citation type="doi">10.1002/pro.5560050824</citation>
  </citations>
</tool>