Skip to Main Content
Prediction of protein secondary structures is an important problem in bioinformatics and has many applications. The recent trend of secondary structure prediction studies is mostly based on the neural network or the support vector machine (SVM). In the paper, a two stage predictor is constructed to predict protein secondary structures. The first stage consists of one predictor based on the support vector machine. Bayesian discrimination is used at the second stage by considering the predicted labels of neighbor residues. The improvement of prediction performances exploits that residues tend to form structures cluster. This method outperforms the predictors based on SVM algorithm alone. Our proposed approach is promising which can be verified by its better prediction performance based on a non-redundant data set.