摘要
提出了一种基于结构特征的蛋白质二级结构预测方法。先对氨基酸的理化特性进行主成分分析,提取出主要影响因素,并融合成3位编码。接着,在原有3位编码基础上加入3位氨基酸在特定二级结构中的倾向因子。编码完成后,使用支持向量机方法进行预测。实验结果表明,改进后的编码方式优于单纯做主成分分析得到的3位编码和5位编码方式,可以有效地用于蛋白质二级结构预测。
This paper provides a method to predict protein secondary structure based on the structuraI characteristics.FirstIy,it extracts the main factors and fuses them into the tribit encoding from physicaI and chemicaI properties of amino acids by principaI component anaIysis.Next,it adds into three propensity factors of amino acids of the specific secondary structure on the originaI basis of coding.After the coding is compIeted,it uses support vector machine to predict the protein secondary structure.
出处
《工业控制计算机》
2015年第4期109-110,113,共3页
Industrial Control Computer
关键词
编码方式
主成分分析
倾向因子
支持向量机
encoding
principaI component anaIysis
propensity factor
support vector machine
