<tool id="rsem_prepare_reference" name="RSEM prepare reference" version="1.1.17">
  <description></description>
  <requirements>
    <requirement type="package" version="1.1.17">rsem</requirement>
    <requirement type="package" version="1.0.0">bowtie</requirement>
  </requirements>
  <command>
    mkdir $reference_file.extra_files_path &amp;&amp; 
    cd $reference_file.extra_files_path &amp;&amp; 
    rsem-prepare-reference 
    #if $polya.polya_use == 'add':
      #if $polya.polya_length:
        --polyA-length $polya.polya_length
      #end if
    #elif $polya.polya_use == 'subset':
      --no-polyA-subset $polya.no_polya_subset
      #if $polya.polya_length:
        --polyA-length $polya.polya_length
      #end if
    #elif $polya.polya_use == 'none':
      --no-polyA
    #end if
    $ntog
    #if $transcript_to_gene_map:
      --transcript-to-gene-map $transcript_to_gene_map
    #end if
    #if $reference.ref_type == 'transcripts':
      $reference.reference_fasta_file
    #else:
      --gtf $reference.gtf
      $reference.reference_fasta_file
    #end if
    $reference_name
  </command>
  <inputs>
    <conditional name="reference">
      <param name="ref_type" type="select" label="Reference transcript source">
        <option value="transcripts">transcript fasta</option>
        <option value="genomic">reference genome and gtf</option>
      </param>
      <when value="transcripts">
        <param name="reference_fasta_file" type="data" format="fasta" label="reference fasta file" 
               help="The files should contain the sequences of transcripts."/>
      </when>
      <when value="genomic">
        <param name="reference_fasta_file" type="data" format="fasta" label="reference fasta file" 
               help="The file should contain the sequence of an entire genome."/>
        <param name="gtf" type="data" format="gtf" label="gtf" 
               help="extract transcript reference sequences using the gene annotations specified in this GTF" />
      </when>
    </conditional>
    <param name="transcript_to_gene_map" type="data" format="tabular" optional="true" label="Map of gene ids to transcript (isoform) ids" >
      <help>
        Each line of should be of the form: gene_id transcript_id ( with the two fields separated by a tab character )
        The map can be obtained from the UCSC table browser  
          group: Genes and Gene Prediction Tracks 
          table: knownIsoforms
        Without a map:
          If a reference genome and gtf is used, then RSEM uses the "gene_id" and "transcript_id" attributes in the GTF file. 
          Otherwise, RSEM assumes that each sequence in the reference sequence files is a separate gene.
      </help>
    </param>
    <param name="reference_name" type="text" value="rsem_ref_name" label="reference name">
      <help>A one work name for this RSEM reference containing only letters, digits, and underscore characters</help>
      <validator type="regex" message="Use only letters, digits, and underscore characters">^\w+$</validator>
    </param>
    <conditional name="polya">
      <param name="polya_use" type="select" label="PolyA ">
        <option value="add" selected="true">Add poly(A) tails to all transcripts</option>
        <option value="subset">Exclude poly(A) tails from selected transcripts</option>
        <option value="none">Do not add poly(A) tails to any transcripts</option>
      </param>
      <when value="add">
        <param name="polya_length" type="integer" value="125" optional="true" label="The length of the poly(A) tails to be added. (Default: 125)">
          <validator type="in_range" message="must be positive " min="1"/>
        </param>
      </when>
      <when value="subset">
        <param name="no_polya_subset" type="data" format="tabular" optional="true" label="List of transcript IDs (one per line) that should should not have polyA tails added."/> 
        <param name="polya_length" type="integer" value="125" optional="true" label="The length of the poly(A) tails to be added. (Default: 125)">
          <validator type="in_range" message="must be positive " min="1"/>
        </param>
      </when>
      <when value="none"/>
    </conditional>
    <param name="ntog" type="boolean" truevalue="--no-ntog" falsevalue="" checked="false" label="Disable the conversion of 'N' characters to 'G' characters in the reference sequences" help="Bowite uses the automatic N to G conversion to to align against all positions in the reference."/>
  </inputs>
  <stdio>
    <exit_code range="1:"  level="fatal" description="Error Running RSEM" />
  </stdio>
  <outputs>
    <data format="rsem_ref" name="reference_file" label="RSEM ${reference_name} reference"/>
  </outputs>
  <tests>
    <test>
      <param name="ref_type" value="genomic"/>
      <param name="reference_fasta_file" value="ref.fasta" ftype="fasta"/>
      <param name="gtf" value="ref.gtf" ftype="gtf"/>
      <param name="reference_name" value="ref"/>
      <output name="rsem_ref">
        <assert_contents>
          <has_text text="ref.grp" />
        </assert_contents>
      </output>
    </test>
  </tests>
  <help>

RSEM HOME PAGE - http://deweylab.biostat.wisc.edu/rsem/

NAME
    rsem-prepare-reference

SYNOPSIS
    rsem-prepare-reference [options] reference_fasta_file(s) reference_name

DESCRIPTION
    The rsem-prepare-reference program extracts/preprocesses the reference sequences and builds Bowtie indices using default parameters. 
    This program is used in conjunction with the 'rsem-calculate-expression' program.

INPUTS
    A fasta file of transcripts
    or
    A genome sequence fasta file and a GTF gene annotation file.  (When using UCSC data, include the related knownIsoforms.txt)

  </help> 
</tool>