Imaging of a molecule. The database will create an "internet for chemists"
Although pharmaceutical companies' spending on research and development has increased more than sixfold in the past decade, producing the next Prozac or Viagra remains an expensive process, based largely on educated guesswork. Developing a new drug requires an average of 15-10 years of development and about 500 million dollars. This is where Barry Beunin and his startup, Libraria, based in San Jose California, come into the picture. "Our goal," says Bionin, "is to create a tool so that scientists can predict which molecules will work well, instead of
use methods of trial and error".
Libraria's online database will contain a kind of internal "cookbook", which will be described
How to synthesize materials. Bionin, who helped develop a technique for mass-producing ingredients called combinatorial chemistry, describes Libraria as a kind of "internet for chemists." Instead of entering search terms to find molecules that can bind to a given protein, scientists will search for them by molecular structure or chemical properties. Like a genomic search engine, the system will offer a list of the most suitable molecules, based on theoretical and empirical findings.
Some experts question whether there is enough molecular information to build the huge database, but Biunin says there are already many types of drugs about which there is enough knowledge to guide chemists in their efforts. "Eventually we will be able to say to a drug researcher, 'Here is the molecule you need to produce, and this is the way to produce it,'" promises Biunin.
All rights reserved to "Haaretz" newspaper 2001
Update - 24/04/2001
https://www.hayadan.org.il/BuildaGate4/general2/data_card.php?Cat=~~~302809348~~~82&SiteName=hayadan
