--- a/rsem_prepare_reference.xml	Thu Dec 05 10:54:28 2013 -0600
+++ b/rsem_prepare_reference.xml	Thu Feb 06 11:46:40 2014 -0600
@@ -32,22 +32,20 @@
     #end if
     $reference_name
   </command>
-
   <inputs>
     <conditional name="reference">
-      <param name="ref_type" type="select" label="">
+      <param name="ref_type" type="select" label="Reference transcript source">
         <option value="transcripts">transcript fasta</option>
         <option value="genomic">reference genome and gtf</option>
       </param>
       <when value="transcripts">
         <param name="reference_fasta_file" type="data" format="fasta" label="reference fasta file" 
                help="The files should contain the sequences of transcripts."/>
-
       </when>
       <when value="genomic">
         <param name="reference_fasta_file" type="data" format="fasta" label="reference fasta file" 
-               help="The files should contain either the sequences of an entire genome."/>
-        <param name="gtf" type="data" format="gtf" optional="true" label="gtf" 
+               help="The file should contain the sequence of an entire genome."/>
+        <param name="gtf" type="data" format="gtf" label="gtf" 
                help="extract transcript reference sequences using the gene annotations specified in this GTF" />
       </when>
     </conditional>
@@ -62,9 +60,10 @@
           Otherwise, RSEM assumes that each sequence in the reference sequence files is a separate gene.
       </help>
     </param>
-    <param name="reference_name" type="text" value="" label="reference name">
+    <param name="reference_name" type="text" value="rsem_ref_name" label="reference name">
+      <help>A one work name for this RSEM reference containing only letters, digits, and underscore characters</help>
+      <validator type="regex" message="Use only letters, digits, and underscore characters">^\w+$</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>
@@ -84,7 +83,6 @@
       </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>
@@ -93,8 +91,15 @@
   <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"/>
+    </test>
+  </tests>
   <help>
-  
 
 RSEM HOME PAGE - http://deweylab.biostat.wisc.edu/rsem/
 
@@ -109,8 +114,9 @@
     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>