Hello Rongkun, I am assuming that you are using the main UCSC Genome browser web site, and not a mirror (http://genome.ucsc.edu/) for the test case.
Test case: Using the main UCSC Browser, it sounds as if the Table browser tool is timing out before your query can complete. This is slightly unexpected for data involving intergenic regions only (no introns). But, to double check, please attempt the procedure for intergenic regions on a single chromosome and see if the result is returned as expected. If so, then the recommendation is to retrieve the data per-chromosome and merge later after download (output from Table browser as a named and compressed .gz file). If this does not work, double check that all queries are based on the same genome assembly (hg19, not hg18) and write back for more troubleshooting advice (include your exact steps - in an email, you can send directly to me - for the failed single-chromosome test query). Files in BED format (similar to genePred, minus a few terminal data fields), can merged and converted to a new data format called "bigBed" which can better accommodate large datasets. This type of data format can be reloaded as a custom track, where the complete genome will be in a single file. Since you are investigating intergenic regions, converting from genePred to BED to bigBED should be result in any data loss. Some links to file formats and to pre-compiled utilities to download for manipulating the data once local in your unix environment: http://genome.ucsc.edu/FAQ/FAQformat http://genome.ucsc.edu/goldenPath/help/hgTracksHelp.html#CustomTracks http://hgdownload.cse.ucsc.edu/downloads.html -> Scroll down to "Source downloads" -> click on Other executable binary utilities are available *here* to find: http://hgdownload.cse.ucsc.edu/admin/exe/ http://genomewiki.cse.ucsc.edu/index.php/Kent_source_utilities (description of utilities in source tree - not all are pre-compiled for download, but all can be retrieved and compiled yourself by obtaining the kent source tree http://genome.ucsc.edu/FAQ/FAQdownloads#download27) If using a mirror, pin-pointing the source of the problem will be more difficult, but not impossible. Start with a check that your code based and data is current and any updates completed without errors (mirrors update genbank tracks, such as RefSeq, weekly, typically on Sundays/Mondays). Confirming this first will eliminate those as variables as a solution is tested. Please let us know if you are able to resolve the issue or provide some more feedback for us to work from, Thank you, Jennifer ------------------------------------------------ Jennifer Jackson UCSC Genome Bioinformatics Group ----- "Rongkun Shen" <[email protected]> wrote: > From: "Rongkun Shen" <[email protected]> > To: [email protected] > Sent: Monday, January 11, 2010 10:38:15 AM GMT -08:00 US/Canada Pacific > Subject: [Genome] custom track of intergenic regions > > Hi, > > I am trying to make IGR (intergenic regions) custom track from refGene > > from hg19 according to the approach shown here: > > https://lists.soe.ucsc.edu/pipermail/genome/2007-March/013036.html > > However, it always ended up with waiting for a long time and with a > broken downloaded .gz file and failing to get custom track in genome > browser. > > Actually I succeeded in hg18 using this approach - the file size is > about 230kb in .gz format. > > I wonder if there is anything I did wrong. > > Thanks, > > --Rongkun > > > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
