Comparison Between Artificial Immune System and other Heuristic Algorithms for Protein Structure Prediction

The search for the global minimum of a potential energy function is very difficult since the number of local minima grows exponentially with the molecule size. We present an Artificial Immune System (AIS) inspired by the clonal selection principle, which has been designed for the protein structure prediction problem (PSP). The proposed AIS employs two special mutation operators, hypermutation and hypermacromutation to allow effective searching, and an aging mechanism which is a new immune inspired operator that is devised to enforce diversity in the population during evolution. When cast as an optimization problem, the PSP can be seen as discovering a protein conformation with minimal energy. Our experimental results demonstrate that the proposed AISis very competitive with the existing state-of-art algorithms for the PSP on lattice models with low computational costs.