PREDICTION OF PROTEIN SECONDARY STRUCTURE BY COMBINING NEAREST-NEIGHBOR ALGORITHMS AND MULTIPLE SEQUENCE ALIGNMENTS

Recently Yi & Lander used a neural network and nearest-neighbor method with a scoring system that combined a sequence-similarity matrix with the local structural environment scoring scheme described by Bowie an d co-workers for predicting protein secondary structure. We have impro ved their scoring system by taking into consideration N and C-terminal positions of alpha-helices and beta-strands and also beta-turns as di stinctive types of secondary structure. Another improvement, which als o decreases the time of computation, is performed by restricting a dat a base with a smaller subset of proteins that are similar with a query sequence. Using multiple sequence alignments rather than single seque nces and a simple jury decision procedure our method reaches a sustain ed overall three-state accuracy of 72.2%, which is better than that ob served for the most accurate multilayered neural-network approach, tes ted on the same data set of 126 non-homologous protein chains.