Artificial neural network model for predicting HIV protease cleavage sites in protein

Knowledge of the polyprotein cleavage sites by HIV protease will refine our understanding of its specificity and be useful for designing specific and efficient HIV protease inhibitors. The search for inhibitors of HIV protease will be greatly expedited if an accurate, robust and rapid method is found for predicting the cleavage sites in proteins by HIV protease. In this paper, a back-propagation model, a typical artificial neural network, is applied to predict the cleavability of oligopeptides by proteases with multiple and extended specificity subsites. HIV-1 protease was selected as a subject of study; 299 oligopeptides were chosen as a training set, and 63 oligopeptides as a test set. Because of its high correct prediction rate (58/63 = 92.06%) and stronger fault-tolerant ability, the neural network method is expected to be a useful technique for finding effective inhibitors of HIV protease, which is one of the targets in designing potential drugs against AIDS. The principle of the artificial neural network method can also be applied to analyzing the specificity of any multi-subsite enzyme. (C) 1998 Elsevier Science Ltd. All rights reserved.