蛋白质折叠预测的禁忌搜索粒子群算法

针对PSO算法晚期收敛速度慢、求解精度差的缺点，提出了一种改进优化算法——将粒子群算法（Particle Swarm Optimization，PSO）与禁忌搜索算法（Tabu Search，TS）结合起来解决基于三维AB非晶格模型的蛋白质折叠预测问题。TS算法的引入提高了粒子群收敛后期的精度，粒子变异机制增强了粒子跳出局部极小值的能力。真实数据实验表明，该算法计算出的蛋白质序列能量值相比其他算法有更高的精确度，能够更好地模拟蛋白质构象，是分析蛋白质结构的一种有效方法。

Tabu search-particle swarm algorithm for protein folding prediction

Considering the problems of slow convergence speed and precision at the late of PSO,the method combining Par-ticle Swarm Optimization（PSO） and Tabu Search（TS） algorithm for the task of predicting protein folding based on the 3D off-lattice model is proposed in this paper.TS is introduced to improve the precision and enhance the capability of jumping out the local minimum through particle mutation mechanism.It is concluded,through the experiment on true data set,that the proposed method outperforms other algorithm on the accuracy of calculating the protein sequence energy value,it is turned to be an effective way to analyze protein structure.