利用混沌差分进化算法预测RNA二级结构

RNA二级结构预测在生物信息学中具有重要意义。本文针对RNA二级结构预测，提出了一种混沌差分进化算法。算法对种群进行混沌初始化，利用混沌扰动产生新的个体，缩小搜索空间；根据个体的适应值和种群密度自适应地对个体进行混沌更新，改善了种群的多样性。该算法充分利用了差分进化算法速度快以及混沌的遍历性、随机性和规律性等特点，有效克服了早熟现象，提高了算法的全局搜索能力。实验证明了算法的有效性。

An Algorithm-based Chaos Differential Evolution for Predicting RNA Secondary Structure

The prediction of RNA secondary structure has important significance in bioinformaties.A chaos differential evolution(CDE)has been proposed for prediction of RNA secondary structure.The basic principle of CDE is that pop- ulation is initialized by chaos,the chaos disturbance is used to get new individuals directly and reduce search space.The individuals is updated by chaos according to individual fitness and its density and improve the diversity of population. The algorithm not only sufficiently exerts the quick speed of differential evolution and chaotic characteristics-random- ness,ergodicity and regularity,but also the global search capability of the algorithm has been enhanced badly ,and the premature of algorithm is avoided effectively.The experiments show that the algorithm is effective.