Sign In

Communications of the ACM

ACM TechNews

Computational Method Predicts New Uses For Existing Medicines


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Atul Butte

Stanford University professor Atul J. Butte can map diseases using data from gene chips that measure which genes are active.

Credit: Peter DaSilva for The New York Times

A recent National Institutes of Health (NIH)-funded computational study analyzing genomic and drug data has been able to predict new uses for existing medicines. "If we can find ways to repurpose drugs that are already approved, we could improve treatments and save both time and money," says NIH's Rochelle M. Long.

The researchers, led by Stanford University's Atul J. Butte, focused on 100 diseases and 164 drugs, creating a computer program to search through the thousands of possible drug-disease combinations to find drugs and diseases whose gene expression patterns essentially canceled each other out.

The researchers noticed that diseases with similar molecular processes clustered together in the analysis, as did drugs with similar effects. By studying unexpected members of these clusters, it could be possible to learn how certain diseases progress and how some drugs work at the molecular level, according to the researchers.

From National Institutes of Health
View Full Article

Abstracts Copyright © 2011 Information Inc. External Link, Bethesda, Maryland, USA 


 

No entries found