Volume learning algorithm artificial neural networks for 3D QSAR studies

The current study introduces a new method, the volume learning algorithm (V LA), for the investigation of three-dimensional quantitative structure-acti vity relationships (QSAR) of chemical compounds. This method incorporates t he advantages of comparative molecular Geld analysis (CoMFA) and artificial neural network approaches. VLA is a combination of supervised and unsuperv ised neural networks applied to solve the same problem. The supervised algo rithm is a feed-forward neural network trained with a back-propagation algo rithm while the unsupervised network is a self-organizing map of Kohonen. T he use of both of these algorithms makes it possible to cluster the input C oMFA field variables and to use only a small number of the most relevant pa rameters to correlate spatial properties of the molecules with their activi ty. The statistical coefficients calculated by the proposed algorithm for c annabimimetic aminoalkyl indoles were comparable to, or improved, in compar ison to the original study using the partial least squares algorithm. The r esults of the algorithm can be visualized and easily interpreted. Thus, VLA is a new convenient tool for three-dimensional QSAR studies.