论文部分内容阅读
提出一种新的组合方法用于β-turns预测和特征分析.该方法包括两步:如何表征β-turns特征和如何构建其预测模型.第一步应用氨基酸广义信息因子分析标度表征蛋白质中β-turns的结构特征,该标度涉及氨基酸的疏水性、α-螺旋与转角倾向、体积性质、构成特征、局部柔性及静电性.第二步以426个蛋白质为训练集样本,通过留1/7法交互验证,基于支持向量机建立β-turns预测模型.该模型分别成功地预测547和823个蛋白的β-turns.所得结果与所对比方法结果相当,更重要的是,SVM模型提供了一些关于β-turns特征的重要结构信息.该组合方法可以进一步尝试用于蛋白质结构预测及特征分析.
A new combinatorial method is proposed for β-turns prediction and feature analysis.The method consists of two steps: how to characterize β-turns and how to construct its prediction model.Firstly, the amino acid generalized information factor analysis β-turns structural features, the scale related to amino acid hydrophobicity, α-helix and corner tendency, volumetric properties, compositional features, local flexibility and electrostatic properties.Second step to 426 protein samples for the training set, by leaving a / 7 method and the β-turns prediction model was established based on support vector machine.The model successfully predicted β-turns of 547 and 823 proteins, respectively.The results obtained are comparable to the results of the comparative method, and more importantly, the SVM model provides Some important structural information about the characteristics of β-turns is also available.This combinatorial method can be further used for protein structure prediction and characterization.