Evolutionary algorithm for protein structure prediction

Abstract

A protein is characterized by its 3D structure, which defines its biological function. The protein structure prediction problem has real-world significance where several diseases are associated with the wrong folding of proteins. Predicting protein structures is computationally intractable. In this paper, we present an improved genetic algorithm (GA) for predicting 3D structures of proteins based on the hydrophobic polar model. We evaluate our proposed GA on a number of protein sequences. The generated protein structures demonstrate improvement over previously published results.