I've run into one bug and a one other smaller issue with the Tophat2 wrapper.

The bug manifests itself when running the tool with "use Defaults" selected results in the following error: cp: cannot stat `/data/galaxy-dev/galaxy-dev/database/job_working_directory/000/281/tophat_out/fusions.out': No such file or directory

It seems that somehow, the filter on the output dataset is not working properly when the value (params['fusion_search']['do_search']) is not explicitly set in the parameters.

The second issue is more minor. It seems that the Tophat2 wrapper assumes that Tophat2 will use Bowtie2 (and not Bowtie). However, the actual Tophat2 program will only use Bowtie2 if it is found in the path. Otherwise it defaults to using bowtie which results in an error (Could not find Bowtie index files) since the wrapper points the tool to the bowtie2 index instead. To make things a bit more robust, I suggest adding bowtie2 as a requirement to the tophat2 wrapper. It also might be a good idea to add versions to those requirements. I've attached a small patch to do that (though it's fairly trivial).

Thanks.

--
Lance Parsons - Scientific Programmer
134 Carl C. Icahn Laboratory
Lewis-Sigler Institute for Integrative Genomics
Princeton University

--- tophat2_wrapper.xml 2012-09-12 11:41:46.000000000 -0400
+++ tophat2_wrapper_withreqs.xml    2012-09-14 13:23:25.000000000 -0400
@@ -3,7 +3,8 @@
     <description>Gapped-read mapper for RNA-seq data</description>
     <version_command>tophat2 --version</version_command>
     <requirements>
-        <requirement type="package">tophat2</requirement>
+        <requirement type="package" version="2.0.0-beta7">bowtie2</requirement>
+        <requirement type="package" version="2.0.4">tophat2</requirement>
     </requirements>
     <command interpreter="python">
         tophat2_wrapper.py
___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/

Reply via email to