Hello,

The whole output line is not included, so it is difficult to tell you exactly what is going on in the case you are looking at, but most likely there is a difference below the gene bound level. Look for an alternate transcript start site or coding region - this should be indicated on the line in one of the other columns. (When not doing gene discovery, these alternate TSS/CDS in your data were not utilized, this is the difference between restricting to only known vs including/discovering novel in the input samples).

Also, just in case you want to confirm some of these (or visualize them) the first column of each file generally has a test id/identifier that can be used to look up what is being testing in the tracking files. For the actual transcripts, going back to through the data in these files to map into the Cuffmerge/Cufflinks transcripts files, where you can review the details. Pulling up the input BAM with the Cufflinks transcripts files, along with the reference annotation, in trackster may help you confirm/decide if you agree with the results or not (you know the coordinates, so can zoom into the region(s)).

The Cuffdiff manual describes each file in detail - this is a great reference. The Galaxy wrapper doesn't alter the output.
http://cufflinks.cbcb.umd.edu/manual.html#cuffdiff_output

Hopefully this helps,

Jen
Galaxy team

On 7/16/13 7:50 AM, Hoang, Thanh wrote:
Hi all,
I am working on RNA-seq using TopHat/Cufflink/Cuffdiff for differential gene expression and new gene discovery ( this is what I am interested in).  However, I found many genes that are repeated in the Cuffdiff's ouput. These are the same genes and at the exact the same locus. There should be only one gene for 1 line. Something like this: 
Genes Locus
Status q1 q2 Log2 Folg change Significance
Lnp 2:74517521-74584544 OK 8.91501 85.2735 3.25779    yes
Lnp 2:74517521-74584544 OK 12.0044 171.352 3.83533    yes

If I re-run the Cuffdiff for differential gene expression only ( No gene discovery), the problem is fixed. Anyone knows how o explain and fix this?
Thank you so much



___________________________________________________________
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/

To search Galaxy mailing lists use the unified search at:

  http://galaxyproject.org/search/mailinglists/

-- 
Jennifer Hillman-Jackson
Galaxy Support and Training
http://galaxyproject.org