基于样本选择的蛋白质关联结构预测  

Prediction method of protein contact based on pattern selection

作　　者：孙鹏飞[1,2] 张健沛[1] 

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001 [2]哈尔滨师范大学计算机科学与信息工程学院,黑龙江哈尔滨150080

出　　处：《计算机与应用化学》2010年第7期937-940,共4页Computers and Applied Chemistry

基　　金：黑龙江省教育厅科技资助项目(11551128)

摘　　要：对于模式识别系统而言,不同的训练样本在建立分类模型时所起的作用不同,以往的蛋白质关联结构预测方法都是从样本集中随机选取一部分样本作为分类器的训练样本,这将降低蛋白质关联结构分类器的预测精度,为改善训练样本对预测精度的影响,本文提出一种基于样本选择及BP神经网络的蛋白质关联结构预测方法。该方法选取与蛋白质关联结构相关的属性进行编码,并采用样本选择技术从编码后的样本集中选取一定的高质量样本构建预测模型,从而有效地对蛋白质关联结构进行预测。本文根据提出的编码方式对从蛋白质数据库PDB中获取的200个蛋白质进行编码,然后用最近邻算法选择训练样本,并使用BP神经网络建立相应的预测模型。实验结果表明,进行训练样本选择能够有效提高蛋白质关联结构的预测精度。The effect of the different training samples is different for the classifier when pattern recognition system is established.The training samples were selected randomly in the past protein contact prediction methods,therefore the prediction accuracy of protein contact was reduced.In order to improve the influence of training samples,a prediction method of protein contact on the basis of pattern selection and BP neural network have been brought forward in this paper.The attributes related with protein contact are extracted and coded in the method and pattern selection is used to select training samples from coded samples in order to improve the precision of protein contact prediction.200 proteins from the PDB database are encoded according to the encoding approach and are taken as models of samples.Then samples are taken on the pattern selection based on the nearest neighbor algorithm and corresponding prediction models are set by using BP neural network.The simulation experiment result indicates that this method of pattern selection can improve the prediction accuracy of protein contact.

关 键 词：蛋白质关联结构 神经网络 样本选择 最近邻算法 

分 类 号：TP39[自动化与计算机技术—计算机应用技术] TQ46[自动化与计算机技术—计算机科学与技术]