基于神经网络蛋白质三级结构预测.pdfVIP

第 36 卷 第 9 期 计 算 机 工 程 2010 年 5 月 Vol.36 No.9 Computer Engineering May 2010 ·人工智能及识别技术· 文章编号：1000—3428(2010)09—0176—02 文献标识码：A 中图分类号：TP181 基于神经网络的蛋白质三级结构预测 1 2 2 蔡娜娜 ，陈月辉 ，李 伟 (1. 济南大学控制科学与工程学院，济南 250022 ；2. 济南大学信息科学与工程学院，济南 250022) 摘 要：在伪氨基酸组成中加入与序列相关的影响因子能够提高蛋白质三级结构预测的准确率。将伪氨基酸组成的特征作为神经网络的输 入，建立分类预测模型。选用粒子群优化算法对神经网络的参数进行优化。分类方法采用一对多的二分类方法。数据集选用 Chou 提出的 204 条蛋白质。实验结果使用 Jackknife 交叉验证，表明该方法能提高预测准确率。 关键词：伪氨基酸组成；粒子群优化算法；Jackknife 交叉验证 Prediction of Protein Tertiary Structure Based on Neural Network 1 2 2 CAI Na-na , CHEN Yue-hui , LI Wei (1. School of Control Science and Engineering, University of Jinan, Jinan 250022; 2. School of Information Science and Engineering, University of Jinan, Jinan 250022) 【Abstract 】Pseudoamino Acid(PseAA) composition can incorporate influence factor of a protein sequence, so as to remarkably enhance the predictive accuracy rate. The feature of PseAA composition is selected as the input of the neural network to make a model of classifying and predicting the protein third structure. Particle Swarm Optimization(PSO) algorithm optimizes the parameters of the neural network. A new classifying method named one vs. others binary classifier is introduced. Two hundred and four protein sequences studied by Chou is used as the dataset. Experimental resu