Using probabilistic neural networks to model the toxicity of chemicals to the fathead minnow (Pimephales promelas): A study based on 865 compounds

We investigate the use of probabilistic neural networks (PNN) to model the acute toxicity (96-hr LC50) to the fathead minnow (Pimephales promelas) bas ed on a 865 chemicals data set. In contrast to most other toxicological mod els, the octanol/water partition coefficient is not used as input parameter . The information fed into the neural network is solely based on simple mol ecular descriptors as can be derived from the chemicals' structures and ind icates the potential of this approach as general methodology for the estima tion of toxicological effects of chemicals. (C) 1999 Published by Elsevier Science Ltd. All rights reserved.