Hello -

The browser/tracks lines used to submit the data were not included in 
your question, so I do not exactly what track attributes you are 
setting. However, from your description, the most likely problem is with 
the autoScale feature. This is "on" by default for custom tracks 
(default is "off" for database track loading, such as a mirror would 
do). If some peaks in the data are much larger in value than others, the 
smaller valued data can graph flat or nearly flat. Your observation that 
limiting the data to smaller regions helps display supports this as the 
problem source.

Track attributes such as windowingFunction, viewLimits, and autoScaling 
can be manipulated to alter display. See a complete discussion of these 
and other attributes here:
http://hgwdev.cse.ucsc.edu/~kent/src/unzipped/hg/makeDb/trackDb/

There are also alternate graphing formats to consider.

http://genomewiki.ucsc.edu/index.php/Selecting_a_graphing_track_data_format

Some suggestions:
http://genome-test.cse.ucsc.edu/goldenPath/help/wiggle.html
http://genome-test.cse.ucsc.edu/goldenPath/help/bedgraph.html
http://genome-test.cse.ucsc.edu/goldenPath/help/bigWig.html

Hopefully this will help you to come up data format/attribute set that 
produces a useful display.

Jennifer

On 3/18/10 6:04 PM, Chen-Yi (Charlie) Chen wrote:
> Hi UCSC genome browser's technical group,
> I am Charlie and I recently used UCSC genome browser to try to visualize
> our ChIP sequencing data, and then I've encountered some problems. We used
> the ht-seq R package that developed by UC Riverside to do the analysis,
> and then output the analysis to a WIG file output. By carefully going
> through the WIG file format the UCSC genome browser supports, I think the
> output WIG file we have is supported by the browser. However, when I
> uploaded the this WIG file to custom track, it doesn't show out any peaks
> on the genome browser, even though the "score" is in the WIG file.
>
> Interestingly, when I excerpt part of my WIG file (ex. I excerpt first
> 10,000 line), then the peaks display just fine.
>
> The following address is our WIG files:
> full file:
> http://radbio.lbl.gov:8080/ChIP/input_chr1.wig
> excerpt file:
> http://radbio.lbl.gov:8080/ChIP/input_chr1_part.wig
>
> Our WIG file is generated from the output of ht-seq package (in R)
> analysis, and the raw data is aligned by Bowtie with UCSC reference genome
> of hg18.
>
> Thanks for your help!
>
> -Charlie-
>
> _______________________________________________
> Genome maillist  -  [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to