Prediction of Transmembrane Helical Segments in Membrane Proteins Using Back Propagation Neural Network with Wavelet Analysis

Transmembrane proteins possess important physiological function. The increasing transmembrane protein sequences from genome projects raise the needs for theoretical methods to predict their structures, especially transmembrane helical segments (TMHs). wnnTM, a new method of BP neural network based on wavelet multiresolution analysis (MRA) is initially developed to predict the number and location of TMHs in membrane proteins. 80 proteins with known 3D-structure randomly selected from Mptopo database are used as test set to evaluate the prediction accuracy of the method. The final results indicate that the proposed method is more effective than BP neural network model only.