使用多特征预测蛋白质棕榈化位点  

Prediction of palmitoylation sites using multiple protein sequence characteristics

作　　者：张田雷[1] 王睿[1] 徐晖[2] 

机构地区：[1]陕西理工学院化学与环境科学学院,陕西汉中723000 [2]汉中市产品质量监督检验所,陕西汉中723000

出　　处：《陕西理工学院学报（自然科学版）》2015年第5期54-59,共6页Journal of Shananxi University of Technology:Natural Science Edition

基　　金：陕西理工学院校级人才启动项目(SLGQD13(2)-4)

摘　　要：采用改进的氨基酸组成、SARAH1疏水尺度值、改进的二肽频率特征、间隔氨基酸对组成特征、蛋白质物理化学性质的自相关函数特征值表征给定的蛋白质序列段,然后用小波频谱来提取特征参数值,用支持向量机来预测棕榈酰化位点。模型查准率为0.880,查全率为0.859,F值为0.869,ROC曲线的面积为0.87。研究结果表明,使用多特征预测蛋白质棕榈化位点方法达到了现有预测算法的水平,能够较准确地预测蛋白质棕榈化位点。Palmitoylation is an important post-translational modification, which participates many cellular processes, including antigen processing, DNA transcription and repair, apoptosis, immune reaction and inflammation, regulating cell surface receptors, ion channels and secretor pathway, nerve and muscle degeneration, viral infections and so on. Hence, the accurate prediction of palmitoylation sites can be of help in under-standing the molecular mechanism of palmitoylation and also in designing various related experiments. Here we present an accurate method to identify palmitoylation sites from protein sequence information using a support vector machine model. It has achieved an accuracy of 88% , which shows that this method will be a useful tool to find palmitoylation sites in a protein.

关 键 词：棕榈化 位点 蛋白质 

分 类 号：O643.322[理学—物理化学]