Hi Tom,

A mismatch between chromosome identifiers can definitely be problematic. These have to be exact.

The TAIR 10 genome in Galaxy has these identifiers:
chr1
chr2
chr3
chr4
chr5
chrM
chrC

Comparing these to your GTF file would be the way to see if any mismatches remain.

On 12/2/12 10:01 AM, Tom Smith wrote:
Hi all,

I've been having trouble running my Arabidopsis thaliana NGS pipeline in galaxy. Specifically, tophat alignment to the built-in-index worked fine (visual assessment in IGV) but I couldn't get cufflinks/Cuffduiff to run using the same annotation.gtf file I'd used in the alignment.
It is curious GTF file worked with Tophat, but then not with Cufflinks/Cuffdiff. Did you really use the Junction/Gene Annotation model option, and the GTF file worked OK with TopHat, then failed later with Cufflinks/diff? This may mean that the problem is not with the GTF file at all, but rather with something else in the method/settings. Did it work with Cuffcompare or Cuffmerge?

There are no known issues with the index and it is the same index being used for all tools (Bowtie/Tophat/Cufflinks-merge-compare-diff), but if you continue to have problems after double checking the IDs and other methods (steps between Tophat and Cufflinks, if any, then steps between Cufflinks and Cuffdiff), please send in a bug report and we can try to provide feedback.
http://wiki.galaxyproject.org/Support#Reporting_tool_errors

Just for reference, this is our tutorial for RNA-seq analysis:
https://main.g2.bx.psu.edu/u/jeremy/p/galaxy-rna-seq-analysis-exercise

Hopefully this helps, but if this does not fully resolve the issue, we can work together through the bug report,

Best,

Jen
Galaxy team

Error message = Error running cufflinks. return code = -11 cufflinks: /lib64/libz.so.1.

Cufflinks/cuffdiff would run without the annotation file and give a differential expression output but without the annotation file the gene_id was meaningless (XLOC_000001 etc) and all I had was a locus for a region with differential expression.

After a lot of frustration I re ran the alignment using a custom built genome and annotation file from Illumina: http://cufflinks.cbcb.umd.edu/igenomes.html  and hey presto, cufflinks/cuffdiff work fine. I did try altering using both chr1 and 1 for chromosome name as a discord between the chromosome naming seemed to frequently be a problem. Apart from this, what other reasons could there be for the built-in-index not working with annotation.gtf files? Is this an issue with the built-in-index (Mouse-ear Cress (Arabidopsis thaliana): Arabidopsis_thaliana_TAIR10)? 

Cheers,

Tom Smith
University of York, UK   


___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

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

-- 
Jennifer Jackson
http://galaxyproject.org