A quantitative structure-activity relationships model for the acute toxicity of substituted benzenes to Tetrahymena pyriformis using Bayesian-regularized neural networks

We have used a new, robust structure-activity mapping technique, a Bayesian -regularized neural network, to develop a quantitative structure-activity r elationships (QSAR) model for the toxicity of 278 substituted benzenes towa rd Tetrahymena pyriformis. The independent variables used in the modeling w ere derived solely from the molecular structure, and the model was tested o n 20% of the data set selected from the whole set by cluster analysis and w hich had not been used in training the network. The results show that the m ethod is robust and reliable and give results for mixed class compounds whi ch are comparable to earlier QSAR work on single-chemical class subsets of the 278 compounds and which employed measured physicochemical parameters as independent variables. Comparisons of Bayesian neural net models with thos e derived by classical PLS analysis showed the superiority of our method. T he method appears to be able to model more diverse chemical classes and mor e than one mechanism of toxicity.