PREDICTING SECONDARY STRUCTURES OF MEMBRANE-PROTEINS WITH NEURAL NETWORKS

Back-propagation, feed-forward neural networks are used to predict the secondary structures of membrane proteins whose structures are known to atomic resolution. These networks are trained on globular proteins and can predict globular protein structures having no homology to thos e of the training set with correlation coefficients (C(i)) of 0.45, 0. 32 and 0.43 for alpha-helix, beta-strand and random coil structures, r espectively. When tested on membrane proteins, neural networks trained on globular proteins do, on average, correctly predict (Q(i)) 62%, 38 % and 69% of the residues in the alpha-helix, beta-strand and random c oil structures. These scores rank higher than those obtained with the currently used statistical methods and are comparable to those obtaine d with the joint approaches tested so far on membrane proteins. The lo wer success score for beta-strand as compared to the other structures suggests that the sample of beta-strand patterns contained in the trai ning set is less representative than those of alpha-helix and random c oil. Our analysis, which includes the effects of the network parameter s and of the structural composition of the training set on the predict ion, shows that regular patterns of secondary structures can be succes sfully extrapolated from globular to membrane proteins.