Neural network prediction of the HIV-1 protease cleavage sites.

A back propagation neural network method has been developed to study the pattern of polypeptides that can be cleaved by the HIV-1 protease. This method can incorporate many characteristics of the peptides, such as hydrophobicity, beta-sheet and alpha-helix propensities. Mutations can also be applied to probe the most important factors that influence the cleavage.