Hi Ying,

You're in luck because I've been working with genome browsers lately, so I think I can help you address your problem. What you're looking for is a visualization of a coverage histogram for the BAM reads produced by Tophat, yes? 

It turns out that some genome browsers provide this automatically as part of their solution for visualizing BAM files b/c BAM files tend to be very large and hence visualizing aggregated data is often the best solution. Both IGV and the Galaxy Trackster Browser support this functionality. I think you'll have to do some simple file conversions to get the display you want in IGV; you can check out the IGV documentation or perhaps Jim can help. I'm not sure if IGB supports this visualization mode for BAM; Ann can chime in with additional information.

The Galaxy Track Browser supports coverage histograms when viewing large regions. When zoomed in, the reads are typically displayed individually, although there is a (very beta) option to create a histogram for the visible set of reads; this option may not work well (yet!) as Tophat reads often have large gaps.

The top track in this visualization shows a coverage histogram for a set of Tophat reads:

http://test.g2.bx.psu.edu/u/jeremy.goecks/v/assembly-of-h1-hesc-rna-seq-data 

Please see my previous email to Vasu for details about setting up a visualization in the Galaxy Track Browser.

Best,
J.


On 4/20/11 5:16 PM, "Ying Zhang" <ying.zhang.yz323@yale.edu> wrote:

Dear Ann and Jeremy:

We have this discussion long time ago, and I am sorry that I brought it
up here
again. I am just thinking that as Ann said, can we add this tool which convert
bam into wig file into galaxy? Or make a workflow to generate a wig
file from a
bam file generate by tophat? In this way we can just easily get a wig
file from
galaxy and will be able to see it in IGB. I know this may seems
unnecessary for
the purpose of statistical analysis, but if we can see the coverage with IGB,
sometimes it is helpful to pick up interesting points quickly for specific
genes. This may seems a old fashion way but my boss is a big fan of using IGB
to see expression file(wig or sgr file) and do some analysis. THanks a lot!

BEst

Ying

Quoting Jeremy Goecks <jeremy.goecks@emory.edu>:

Hi all,

Ann is correct - Tophat does not produce .wig files when run anymore.
However, it's fairly easy to use Galaxy to make a wiggle-like
coverage file from a BAM file:

(a) run the pileup tool on your BAM to create a pileup file;
(b) cut columns 1 and 4 to get your coverage file.

A final note: it's often difficult to visualize coverage files
because they're so large. You might be better off visualizing the BAM
file and using the coverage file for statistics.

Best,
J.

Hello,

I think I know the answer (sort of) to this question.

This may be because newer versions of tophat stopped running the "wiggles"
program, which is still part of the tophat distribution and is the program
that makes the "coverage.wig" file.

A later version of tophat might bring this back, however - there's a note to
this effect in the tophat python code.

So if you can run wiggles, you can make the "coverage.wig" file on your own.

A student here at UNC Charlotte (Adam Baxter) made a few changes to the
"wiggles" source code that would allow you to use it with samtools to make a
"coverage.wig" file from the "accepted_hits.bam" file that TopHat creates.

If you (or anyone else) would like a copy, please email Adam, who is cc'ed
on this email.

We would be happy to help add it to Galaxy if this would be of interest to
you or other Galaxy users.

If there is any way we can be of assistance, please let us know!

Very best wishes,

Ann Loraine


On 2/21/11 3:39 PM, "Ying Zhang" <ying.zhang.yz323@yale.edu> wrote:

Hi:

I am using tophat in galaxy to analyze my paired-end RNA-seq data
and find out
that after the tophat analysis, we can not get the wig file from it anymore
which is used to be able to. Do you have any idea of how to still
be able to
get the wig file after tophat analysis? Thanks a lot!

Best

Ying Zhang, M.D., Ph.D.
Postdoctoral Associate
Department of Genetics,
Yale University School of Medicine
300 Cedar Street,S320
New Haven, CT 06519
Tel: (203)737-2616
Fax: (203)737-2286
_______________________________________________
The Galaxy User list should be used for the discussion
of Galaxy analysis and other features on the public
server at usegalaxy.org. 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/

--
Ann Loraine
Associate Professor
Dept. of Bioinformatics and Genomics, UNCC
North Carolina Research Campus
600 Laureate Way
Kannapolis, NC 28081
704-250-5750
www.transvar.org


_______________________________________________
The Galaxy User list should be used for the discussion
of Galaxy analysis and other features on the public
server at usegalaxy.org. 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/




Ying Zhang, M.D., Ph.D.
Postdoctoral Associate
Department of Genetics,
Yale University School of Medicine
300 Cedar Street,S320
New Haven, CT 06519
Tel: (203)737-2616
Fax: (203)737-2286

--
Ann Loraine
Associate Professor
Dept. of Bioinformatics and Genomics, UNCC
North Carolina Research Campus
600 Laureate Way
Kannapolis, NC 28081
704-250-5750
www.transvar.org