Neural network applications in chemistry begin to appear

DALLAS Predictions of product distributions from chemical reactions, frequencies of adverse drug reactions, and three-dimensional structures of proteins from amino acid sequence are among chemical applications emerging from the little known but rapidly growing field of neurocomputing. According to speakers at a symposium on neural networks held by the Division of Computers in Chemistry, interest in the use of neural systems to solve chemical problems is beginning to increase. "There haven't been any meetings before this on chemical applications of neural networks, as far as I know," says Mark E. Lacy of Norwich Eaton Pharmaceuticals, who organized and presided over the symposium. "It was a bear to track down people across the country who were not only working in this area, because it's so new, but [who] also were allowed to say something about it, because it's frequently proprietary. People are just starting to get into this area. It's really on the cutting edge of ...