An Ecology-Based Evolutionary Algorithm Applied to the 2D-AB Off-Lattice Protein Structure Prediction Problem

This paper applies an ecology-inspired algorithm (ECO) to solve a complex problem from bioinformatics. The ecological-inspired algorithm represents a new perspective to develop cooperative evolutionary algorithms. Different algorithms are applied to compose the computational ecosystem, both homogeneously and heterogeneously. The aim is to search low energy conformations for the Protein Structure Prediction problem, concerning the 2D-AB off-lattice model. From the results, the heterogeneous configuration obtained the best conformations for almost all cases, possibly due to the use of different intensification and diversification strategies provided by different search algorithms.

I. Introduction

Many Bioinformatics problems are featured mainly to be non-linear and strongly constrained. This is the case of the protein structure prediction problem approached in this paper. Due to the limitations of exact methods for solving such a class of problems, the need for more robust techniques arises. Along decades, Evolutionary Computation (EC) and Swarm Intelligence (SI) have provided a large range of flexible and robust optimization methods, capable of dealing successfully with complex optimization problems. Both EC and SI provide population-based methods where each individual of a population represents a tentative solution to the problem to be solved. With such diversity of search strategies [1], it is possible to establish an analogy with the dynamics of biological ecosystems.