Hello again Damien, I realized that some of what I said was misleading or confusing.
First I want to clarify the difference between sessions saved in files and "named sessions" created with a genomewiki login. With sessions saved in files, all of the settings you have changed in the Genome Browser (your "cart") are saved to a file, which you or anyone you share the file with can upload at any time and see the same settings. The disadvantage of the file, as you have noticed, is that it doesn't actually save the custom tracks that were loaded when the session file was made, so the custom tracks expire. When you create a "named session" (as described here: http://genome.ucsc.edu/goldenPath/help/hgSessionHelp.html#Create), and the session includes custom tracks, the tracks do not expire for 4 months from the last time they are accessed (at least, we make our best effort to keep them that long). That said, there are some different ways you could go about sharing the data in your lab: (1) load the custom tracks as you have been doing but create a named session instead of a session file, so the custom tracks will expire in 4 months instead of 48 hours. Then, instead of sharing the file, share the link to your named session (or your username and the session name). (2) do not use a session at all, and share your custom tracks via a URL as outlined here: http://genome.ucsc.edu/goldenPath/help/customTrack.html#SHARE) It is possible to construct a single URL to load several custom tracks at once: put each custom track data file in a web-accessible location, then make another file listing the URLs for each data file, and use the URL of that file in a link-to-share, which will look something like this: http://genome.ucsc.edu/cgi-bin/hgTracks?org=organism_name&; position=chr_position&hgt.customText=URL This method will cause your custom tracks to be loaded each time someone opens the Genome Browser with the URL you construct. Other cart settings will not be shared. (3) do the same thing as in #2, except first convert the files to bigBeds. The track lines need to use the "bigDataUrl=" syntax, outlined here: http://genome.ucsc.edu/goldenPath/help/bigBed.html This has the added advantage that only portions of files that are viewed are uploaded to UCSC, so there is no need to wait for the whole file to upload before viewing data. Once any portion of a bigBed file is uploaded, the data is cached at UCSC for several hours, so anyone looking at a custom track that references the same bigBed file will be able to see that data without waiting. (4) do the same thing as in #1 (save a named Session), but convert the files to bigBeds first. With this method you get the advantages of the bigBed files, plus you would be able to share a customized view of tracks (because you would be sharing your cart settings, too). Sorry to be so long-winded! I hope this helps you decide on the best option for your lab. -- Brooke Rhead UCSC Genome Bioinformatics Group On 04/30/10 12:07, Brooke Rhead wrote: > Hi Damien, > > Custom tracks saved in session files have the typical 48 hour lifespan. > However, custom tracks saved in sessions using a genomewiki login last > for up to 4 months after the last time the session is accessed. (Note > that we don't guarantee the 4-month custom track life spans; system > failures might cause data loss before that date, so be sure to save > custom tracks and be prepared to reload them and re-save the session.) > > There is another option entirely, however. If you convert your BED data > into the bigBed file format (described here: > http://hgwdev.cse.ucsc.edu/goldenPath/help/bigBed.html) and put it in a > web-accessible location for the Genome Browser to display, nothing will > expire. Furthermore, only the portions of the file viewed by users on > your end will be uploaded to UCSC at all (and once a portion is > uploaded, it is saved here for subsequent access), so you will not need > to deal with constant loading and reloading of data. > > The bigBed file format is relatively new, and we recommend using it over > saving a session in genomewiki. The link above should contain all of > the instructions for setting up a bigBed custom track, but please feel > free to contact the list again if you have further questions. > > -- > Brooke Rhead > UCSC Genome Bioinformatics Group > > > On 04/30/10 03:11, Damien Bruno wrote: >> Dear UCSC Team, >> >> As part of our routine diagnostic work our laboratory relies heavily >> on the UCSC Genome Browser for interpretation of whole-genome >> microarray data. This involves the use of several custom tracks >> (loaded as bed files) that have been developed by us as well as >> others in the 'Cytogenetics' community. To avoid repetitive loading >> of these bed files by multiple users we recently transitioned to >> using a session file. >> >> Although this has simplified the process (for multiple users), the >> custom tracks contained within the session have the typical lifespan >> of 48 hours. Is it possible to extend the lifespan of the custom >> tracks used in this session? >> >> Yours faithfully, >> >> Damien Bruno >> >> >> Damien Bruno, PhD Medical Scientist Cytogenetics Department, VCGS >> Pathology, Murdoch Childrens Research Institute Flemington Road, >> Parkville 3052 Australia Tel: 61-3-8341 6258 Email: >> [email protected] >> >> _______________________________________________ Genome maillist - >> [email protected] >> https://lists.soe.ucsc.edu/mailman/listinfo/genome > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
