计算机与数字工程2017，Vol.45Issue(11)：2093-2096,2117,5.DOI:10.3969/j.issn.1672-9722.2017.11.003
基于特征相关性选择的二硫键预测算法
Predicting Disulfide Connectivity Based on Correlation Coefficients Selection
摘要
Abstract
Disulfide connectivity is one of significant protein structural characteristic. Previous prediction methods usually used support vector regression,which didn 't consider the correlation between different features. According to traditional prediction methods,based on fisher score,this paper calculated correlation coefficient of each pair of features after feature selection,then de-leted the features with high correlation coefficient. Based on the rest features,support vector regression was used to train model and test. 4-fold validation was used on our benchmark dataset to gain a hopeful result comparing with previous results.关键词
生物信息学/二硫键/支持向量回归/相关系数/特征选择Key words
bioinformatics/disulfide bond/support vector regression/correlation coefficient/feature selection分类
信息技术与安全科学引用本文复制引用
刘坤..基于特征相关性选择的二硫键预测算法[J].计算机与数字工程,2017,45(11):2093-2096,2117,5.基金项目
国家自然科学基金项目(编号:61373062,61371040)资助. （编号:61373062,61371040）
