DATA MINING SHOWS NEW PROMISE IN EFFECTIVE DRUG DEVELOPMENT

Posted on February 14, 2011 by Jeffrey Newman

A robot/computer known as Eve, using articifial intelligence and data mining to analyse the results of pharmacological experiments, is making informed predictions has to the effectiveness of of drug treatments. Eve relates the chemical structures of different chemical compounds to their pharmacological activity to learn which compounds should be tested based on which are most likely to be effective targets, based on her analysis of past experiments. The work is being done by researchers at the Jozef Stefan Institute in Ljubljana. Currently when a new drug is sought, researchers conduct blilnd studies of hundreds of thousands of compounds applying them to an assay for a disease. The process used by Eve may save time and money in finding more effective compounds sooner. From experimental data collected by Eve, patterns emerge which can be used for her to make informed guesses about what compounds would be effective and which would not. The hope is to reduce the time it takes to develop new drugs.