Multiple Feature Fusion Protein Tertiary Structure Prediction

Predicting protein tertiary structure from its primary amino acid sequence is a challenging mission for bioinformatics. In this paper we put forward a novel approach for predicting the tertiary structure of protein and construct an Error Correcting Output Codes(ECOC) classification model on the basis of Particle swarm optimization(PSO) and neural network(NN). Three feature extraction methods, which are Amino Acid Composition, Amino Acid Frequency and Hydrophobic Amino Acid Combination, respectively, are employed to extract the features of protein sequences. To evaluate the efficiency of the proposed method we choose a benchmark protein sequence dataset (640 dataset) as the test data set. The final results show that our method is efficient for protein structure prediction.