A supervised artificial neural network model has been developed for the accurate prediction of trypanocidal activity of any chemical compound. The artificial neural networks (ANNs) approach was used for the key steps of the development of a predictable model: descriptors generation using the CODES® program, dimension reduction strategy and model developing by means of supervised back propagation neural network.
The definition of the molecules was achieved from a non-supervised neural network using a new methodology, the CODES® program. (This program codifies each molecule into a set of numerical parameters taking into account exclusively the information of its chemical structure)