摘要
蛋白质的功能常体现在生物大分子的相互作用中,识别蛋白质相互作用位点对于研究蛋白质功能发挥着重要作用。蛋白质间主要通过表面残基发生相互作用,蛋白质相互作用形成复合体时,只有部分表面残基参与了该过程。基于序列谱信息,提取序列上相邻残基的序列谱作为输入特征向量,对大小为3和7的残基信息窗(win3,win7),分别采用支持向量机(SVM)分类器对蛋白质相互作用位点进行预测、比较和分析。最终实验结果为:win3的平均正确率为69.31%,win7的平均正确率为69.68%。
Protein usually represents its function through interactions among biological molecules. Identifying protein-protein interaction sites plays an important role in protein's function. The interactions among proteins are produced by surface residues mainly, and only part of surface residues participate in this produce. Adjacent residue sequences profile are as input vectors of two information windows of residue (win3 ,win7) and interaction sites are classified by support vector machine (SVM). The result shows that the average accuracy of win3 is 69.31%, and the average accuracy of win7 is 69.68%.
出处
《计算机技术与发展》
2008年第10期127-129,132,共4页
Computer Technology and Development
基金
安徽省自然科学基金(KJ2007B239)
安徽建筑工业学院青年科研基金(200510304)
安徽省高校青年教师科研资助计划(2007jql140)
关键词
蛋白质相互作用
序列谱
残基信息窗
支持向量机
protein-protein interaction
sequence profile
information window of residue
SVM