Discrete Binary Adaptive Bat Algorithm for RNA Secondary Structure Prediction

摘要

RNA secondary structure is one important problem in bioinformatics. In this paper, one discrete binary adaptive bat algorithm is designed to solve it. In the standard version, pulse rate is exponential increased significantly, and then, pulse rate keeps a constant in most generations, this may result a fix selection pressure. To provide a large selection pressure, a linearly dynamic pulse rate selection strategy is designed. Furthermore, due to the special requirements of RNA secondary structure problem, Sigmoid function is also employed to determine the binary-value of each feature. Ten sequences from the comparative RNA website were selected for the evaluation of the proposed method. Simulation results show adaptive bat algorithm is better than Mfold.