I'd like to extract, for a rather long list of... - The Life Scientists

FriendFeed is the easiest way to share online. Learn more »

I'd like to extract, for a rather long list of human genes, info on tissues or cell types where they are known to be important in some automated way from literature. Any suggestions for where to start? Extracting Gene Ontology terms will miss many known gene-tissue associations. Is there a text mining tool that can find gene/cell-type associations?

October 31 - Comment - Like - Share

Andreas Matern, Carl Fulp, Nakao M. and 3 other people liked this

Only starting point that I can suggest is Acembly: see for instance http://www.ncbi.nlm.nih.gov/IEB... - then click "expression tissue" in menu bar at top of page. I don't know how easy it is to mine - there are some downloads available. NCBI GEO Profiles might be another entry point. There are sure to be tissue expression databases but I don't know what they are. - Neil Saunders

Now I think some more, BioGPS (successor to SymAtlas) is also a good start: http://biogps.gnf.org - Neil Saunders

Thanks Neil, AceView is a nice resource that I had forgotten about. I think I can get started by digging there. BioGPS I had already been playing around with. It is also a great resource, but I would like to collect literature mentions of genes in the context of tissue specific activity and/or diseases in addition to expression data. - Mikael Huss

You may want to have a look at HPRD (Human Protein Reference Database) http://www.hprd.org/ . The database is reporting gene - tissue information based on extensive literature curation. - Khader Shameer

Thanks, Khader, another good suggestion. BTW, http://www.proteinatlas.org/ is a nifty resource, but it does not seem to provide any kind of downloads. - Mikael Huss

I'm not sure if there are Genetics/Celluar biology professionals on http://sciencestack.com yet, but you might want to try to ask there. - Jane Breezler