A memetic algorithm for protein structure prediction in a SD-lattice HP model

Abstract

This paper presents a memetic algorithm with self-adaptive local search, applied to protein structure prediction in an HP, cubiclattice model. Besides describing in detail how the algorithm works, we report experimental results that justify important implementation choices, such as the introduction of speciation mechanisms and the extensive application of local search. Test runs on 48-mer chains show that the proposed algorithm has promising search capabilities. © Springer-Verlag 2004.