多种信息融合的细胞凋亡蛋白质的亚细胞定位预测  

Prediction of Apoptosis Protein Subcellular Localization Based on Hybrid Feature Parameters

作　　者：薛济先 陈颖丽[1] 翟媛媛 

机构地区：[1]内蒙古大学物理科学与技术学院,内蒙古呼和浩特

出　　处：《计算生物学》2016年第3期62-71,共10页Hans Journal of Computational Biology

基　　金：国家自然科学基金(61361015);教育部第46批留学回国人员科研启动基金的支持。

摘　　要：研究表明mRNA的序列和结构特性与蛋白质的亚细胞定位有一定关系。本文提取了细胞凋亡蛋白质的两种mRNA信息:三阅读框3-mer mRNA序列频数信息、mRNA二级结构-序列模式信息,并结合细胞凋亡蛋白质的氨基酸物理化学性质、氨基酸黏性特征和进化信息,构成特征向量来表示mRNA和蛋白质序列,利用支持向量机算法,对四种不同亚细胞位置的细胞凋亡蛋白质进行预测。研究发现融合mRNA信息与氨基酸信息后预测效果更佳,在Jackknife检验下,预测总精度达到82.18%,且独立测试集预测总精度达到78.26%。结果表明,mRNA的序列和结构特性有助于细胞凋亡蛋白质的亚细胞定位预测。Studies have shown that sequence and structure characteristics of the mRNA have a certain re-levance with subcellular localization of protein. In this article, it extracted two mRNA information of apoptosis proteins: the three reading frame 3-mer mRNA sequence frequency information and mRNA secondary structure-sequence mode information, and to construct feature vector which indicate mRNA and amino acid sequence with physicochemical properties, stickiness and evolutionary information of apoptosis proteins. Meanwhile, by using support vector machine algorithm, apoptosis proteins of four different subcellular localizations were predicted. The study found that the hybrid of mRNA and AAs information promoted prediction result, and the overall prediction access rate achieved 82.18% while 78.26% for independent test datasets by the Jackknife test. Prediction results show that sequence and structure characteristics of the mRNA contribute to prediction of the subcellular localization of apoptosis proteins.

关 键 词：细胞凋亡蛋白 MRNA二级结构 亚细胞定位 

分 类 号：R73[医药卫生—肿瘤]