MIT Computer Scientist Creates Algorithms for Better Drug Designs

Posted on September 1, 2010 by Jeffrey Newman

Shivani Agarwal, a Postdoc student in the Computer Science Dept at Harvard has created a new kind of algorithm that can predict a drugs’ success more reliably than those in current use. The difference is in how the algorithms measure the accuracy of the prediction about drugs. Traditional algorithms when predicting the accuracy of a drug, look at each score for a drug separately and ask whether it will reflect a drugs success or failure. The new formulas being created looks at scores in pairs and asks whether the function got their order right. See anking Chemical Structures for Drug Discovery: A New Machine Learning Approach

AbstractFull Text HTMLHi-Res PDF[1082 KB]PDF w/ Links[352 KB]
Shivani Agarwal*†, Deepak Dugar‡ and Shiladitya Sengupta§
Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, and Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115, USA
J. Chem. Inf. Model., 2010, 50 (5), pp 716–731
DOI: 10.1021/ci9003865
Publication Date (Web): April 13, 2010
Copyright © 2010 American Chemical Society