A fuzzy adaptive particle swarm optimization for RNA secondary structure prediction

The secondary structure prediction of RNAs is an important classical problem in bioinformatics. The standard solution is to predict the secondary structure possessing the minimum free energy. In this paper, we propose a fuzzy adaptive particle swarm optimization (FPSO) combining particle swarm optimization (PSO) and fuzzy logic control (FLC) to predict RNA secondary structure with the minimum energy. The proposed method aims to predict pseudoknots in the large search spaces. The numerical results and statistical analysis show that the proposed approach is capable of finding an optimal feature subset from a large noisy data set. The performance of the proposed method is compared with that of the PSO based, genetic algorithm (GA) based, simulated annealing based (SA) and ant colony optimization (ACO) based methods on five sequences from the comparative RNA website. The results show that the prediction accuracy rate is significantly better than that of the other methods with minimum energy.