基于激光光谱分析的RNA二级结构预测  

Secondary structure prediction of RNA based on laser spectral analysis

作　　者：李熹 杨运煌[2] LI Xi;YANG Yunhuang(College of Software and Information Security,Guangxi University for Nationalities,NanningGuangxi 530006,China;Wuhan Institute of Physics and Mathematics,Chinese Academy of Sciences,Wuhan Hubei 430071,China)

机构地区：[1]广西民族大学软件与信息安全学院,南宁530006 [2]中国科学院武汉物理与数学研究所,武汉430071

出　　处：《激光杂志》2019年第6期32-36,共5页Laser Journal

基　　金：广西高校中青年教师基础能力提升项目(No.2017KY0177)

摘　　要：基于副本交换的局部增强差分进化蛋白质结构预测方法,对RNA二级结构进行预测时,信息的筛选不够充分,预测性能差。提出基于激光光谱分析的RNA二级结构预测方法,采用基于激光光谱能量的动态划分算法搜索光谱能量最细微的光谱茎区,通过Hopfield网络算法预测RNA二级结构时,采用贪心算法获取待测RNA光谱茎区池,使用动态规则方法对RNA双光谱序列里光谱能量最细微的RNA光谱茎区和光谱茎区池区间隔进行比照,间隔过长的光谱茎区通过Hopfield神经网络的运算“变短”从而减少干扰,RNA光谱茎区池里的光谱茎区用其预测值进行更正,实现RNA光谱茎区池的改进,再通过贪心算法和Hopfield神经网络基于改进后的RNA光谱茎区池对RNA二级结构进行预测。实验表明,所提方法对RNA二级结果预测性能的最高均值是96.98%,平均预测时间为6.2s,预测性能高。Based on the partial enhanced differential evolutionary protein structure prediction method based on copy exchange,when the RNA secondary structure is predicted,the information screening is insufficient and the prediction performance is poor.A method for the prediction of RNA secondary structure based on laser spectral analysis is proposed.A dynamic partition algorithm based on laser spectral energy is used to search for the smallest spectral region of spectral energy.When predicting the secondary structure of RNA through Hopfield network algorithm,a greedy calculation is used to obtain the stem pool of the RNA to be measured.Using the dynamic rule method to compare the interval between the RNA spectral stem region with the smallest spectral energy in the RNA bispectral sequence and the spectral stem region,the stem area with long spectral is shortened through the operation of the Hopfield neural network to reduce interference.The spectral stem area in the pool of RNA spectroscopy is corrected with its predicted value to improve the RNA spectral stem pool,and then the RNA secondary structure is predicted by the greedy algorithm and the Hopfield neural network based on the improved RNA spectral stem pool.Experiments show that the highest mean of the proposed method is 96.98%,and the average prediction time is 6.2s.

关 键 词：激光光谱 RNA二级结构 贪心算法 光谱茎区 光谱能量 

分 类 号：TN209[电子电信—物理电子学]