| Natural creatures sometimes behave as a swarm. Swarm intelligence algorithms utilizes these behavior in establishing models. There are many swarm intelligent optimization, particle swarm algorithm (PSO) is one of that algorithms, which has been successfully applied to many optimization problems and shown its high search speed in these applications.However, as the dimension and the number of local optima of problems increase, PSO is easily trapped in local optima.Protein is an important component of living organisms, which is the main bearer of life activities. The structure of protein determines its function in molecular. Predicting the structure of protein through its sequence of amino acids is a complex and challenging problem in computational biology. The native structure of a protein is associated with the structure of the global minimum of the free energy consisting of the intra-molecular interaction among protein atoms and between the proteins and surrounding solvent molecules. Based on this minimum free energy theory, many simplified protein models have been proposed to predict the structure of protein. Toy model is one of the simplest and most effective protein models proposed by Stillinger. Though toy model is one of the simplest and effective models, it is still require very extensive computation to predict its structure as the increase of amino acids. So it is difficult to predict the structure rapidly.In this paper, have proposed an improved PSO algorithm, called EPSO (Euclidean Particle Swarm Optimization), which has greatly improved the ability of escaping form local optima, and it has confirmed its excellent performance in benchmark functions. Then applied the new algorithm to the structure prediction of toy model both on artificial and real protein sequences and compared with the results reported in other literatures. The experimental results demonstrated that the new algorithm was efficient in protein structure prediction problem in toy model.
|
PDF Full Text Request