Validation of counter propagation neural network models for predictive toxicology according to the OECD principles: A Case study

OECD proposed five principles for validation of QSAR models used for regulatory purposes. We present a case study in Kohonen neural networks and counter propagation neural networks, which we have investigated whether these principles can be applied to models based on. The study is based on a counter propagation network built on toxicity of 541 compounds to the fish fathead minnow, and shows that most, if not all, of the OECD criteria can be met in modeling using this network approach.