多特征融合提取算法结合支持向量机预测膜蛋白类型.pdfVIP

43 7 上 海 交 通 大 学 学 报 Vol. 43 No. 7 2009 7 JOU RNAL OF SH ANGHAI JIAOT ONG UNIVERSITY Jul. 2009 : 100 -24 7( 2009) 07- 1172-05+ 1179 1 1 1 1, 2 姜 彬 , 王正华 , 王勇献 , 贺细平 ( 1. , 4 10073; 2. , 410128) : 针对膜蛋白分类预测问题, 在氨基 组分基础上引入氨基 位置信息, 计算多种氨基 残 基指数的相关系数并选择最优组合方式; 融合2类特征信息对膜蛋白序列进行特征提取; 采用支持 向量机算法作为分类器, 构建了一种新型膜蛋白分类模型, 在自检验Jackknife 检验和独立测试集 检验3 种典型方式下, 预测准确率分别为98. 25%88. 10%和95. 2% . 结果表明, 多特征融合能够 有效提取膜蛋白序列的特征信息, 与现有方法相比, 该分类模型具有较高的分类预测成功率. : 膜蛋白; 权重氨基 组成; 氨基 指数; 相关系数; 支持向量机 : TP 39 : A Predict ion of Membrane Protein Types by Combining Feat ures and Weighted-Support Vector Machine 1 1 1 1, 2 J IA N G Bin , WA N G Zheng-hua , WA N G Yong-x ian , H E X i-p ing ( 1. School of Computer, National U niversity of Defense Technology, Changsha 410073, China; 2. School of Information Science and Technology , Hunan Agricultural University, Changsha 410128, China) Abstract: M embrane protein plays a crucial role in cells and makes the material basis for cells to implement various functions. In order to predict the type of membrane protein, w hich is a crucial fundamental re- search in the field of the structure and function of membrane protein, this paper introduced the position of amino acid and calculated mult-i amino acid index correlation coefficients. Then it constructed a new type of membrane protein classification model that combines tw o feature classes and support vector machine ( SVM) . U nder three typical methods ( self-consistency, Jackknife and independent dataset) , the accuracy rate of the prediction is 98. 25%, 88. 10% and 95. 2% respectively . The experimental results demo