Evolutionary Neuro-Fuzzy System for Protein Secondary Structure Prediction

Abstract

Protein secondary structure prediction is an essential step for the understanding of both the mechanisms of folding and the biological function of proteins. Experimental evidences show that the native conformation of a protein is coded within its primary structure. This work investigates the benefits of combining genetic algorithms, fuzzy logic, and neural networks into a hybrid Evolutionary Neuro-Fuzzy System, especially for predicting a protein’s secondary structure directly from its primary structure. The proposed system will include more biological information such as protein structural class, solvent accessibility, hydrophobicity and physicochemical properties of amino acid residues to improve accuracy of protein secondary structure prediction. The proposed system will experiment on three-class secondary structure prediction of proteins, that is, alpha helix, beta sheet or coil. The experimental results indicate that the proposed method has the advantages of high precision, good generalization, and comprehensibility. The method also exhibits the property of rapid convergence in fuzzy rule generation.