PROTEIN SECONDARY STRUCTURE PREDICTION OF QscR PROTEIN USING FFA OPTIMIZED ANN

Abstract

Identifying Secondary structure of protein becomes more important in designing drugs. Presently, machine learning algorithms like Artificial Neural Network (ANN) and Support Vector Machine (SVM) has been utilized to identify the Protein Secondary Structure (PSS). Over the years, many hybrid methods are evolved for the optimization. In this work, to train a FFNN, a heuristic nature inspired algorithm namely firefly algorithm (FFA) is incorporated with BP algorithm to get a quick and enhanced convergence rate in training FFNN. Using this algorithm, enhanced convergence within a very few repetitions can be attained. It has also been observed that by this approach, higher accuracy in Protein Secondary Structure (PSS) of QscR can be achieved when compared to that of other existing techniques.