Use of statistical and neural net approaches in predicting toxicity of chemicals

Hierarchical quantitative structure-activity relationships (H-QSAR) have be en developed as a new approach in constructing models for estimating physic ochemical, biomedicinal, and toxicological properties of interest. This app roach uses increasingly more complex molecular descriptors in a graduated a pproach to model building. In this study, statistical and neural network me thods have been applied to the development of II-QSAR models fur estimating the acute aquatic toxicity (LC50) of 69 benzene derivatives to Pimephales promelas (fathead minnow). Topostructural, topochemical, geometrical, and q uantum chemical indices were used as the four levels of the hierarchical me thod. It is clear from both the statistical and neural network models that topostructural indices alone cannot adequately model this set of congeneric chemicals. Not surprisingly, topochemical indices greatly increase the pre dictive power of both statistical and neural network models. Quantum chemic al indices also add significantly to the modeling of this set of acute aqua tic toxicity data.