基于ACC变换和RFE算法的蛋白质亚核定位预测  

Predicting protein subnuclear location using ACC transformation and RFE algorithm

作　　者：李小苇 刘太岗[2] 陶珮莹 王春华[2] LI Xiaowei;LIU Taigang;TAO Peiying;WANG Chunhua(College of Food Science & Technology, Shanghai Ocean University, Shanghai 201306, China;College of Information Technology, Shanghai Ocean University, Shanghai 201306, China)

机构地区：[1]上海海洋大学食品学院,上海201306 [2]上海海洋大学信息学院,上海201306

出　　处：《计算机工程与应用》2016年第15期83-87,共5页Computer Engineering and Applications

基　　金：上海市教育委员会科研创新项目(No.13YZ098);上海高校青年教师培养资助计划(No.ZZhy12028);上海海洋大学博士科研启动基金

摘　　要：获取真核细胞中细胞核内蛋白质定位的信息对注解蛋白质功能具有非常重要的意义。针对于利用计算方法预测蛋白质在亚核水平上的定位更具挑战性的问题,提出了基于自互协方差变换与递归特征消除预测蛋白质亚核定位的方法。该方法基于位置特异性得分矩阵利用自互协方差变换构建蛋白质序列的特征向量,采用递归特征消除法进行特征选择,选用支持向量机作为预测工具,并在两个经典数据集SC714和LD504上进行了夹克刀测试。实验结果表明,该方法比大多数已报道的预测方法具有更高的预测准确率。The knowledge of protein subnuclear location in eukaryotic cells plays a very important role for understandingthe biological functions of proteins. As it is very difficult and challenging to predict it at the subnuclear level using computationalmethods, a method which combines Auto Cross Covariance(ACC)transformation and Recursive Feature Elimination(RFE)has been proposed. ACC transformation is first employed to extract features to represent the proteins based onPosition Specific Scoring Matrix(PSSM). Then, RFE is adopted to select the optimal features. Finally, the reduced featuresare input to a Support Vector Machine(SVM)to perform the prediction. Jackknife tests on two widely used datasets(SC714 and LD504)show that the proposed method is very promising and performs better than most of existing methods.

关 键 词：蛋白质亚核定位 位置特异性得分矩阵 自互协方差变换 递归特征消除 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] Q51[自动化与计算机技术—控制科学与工程]