中文摘要：

蛋白质折叠类型识别是蛋白质结构研究的重要内容,折叠类型分类是折叠识别的基础。通过对ASTRAL-1.65数据库α类蛋白质所属折叠类型进行系统研究,建立蛋白质折叠类型模板数据库,提取反映折叠类型拓扑结构的模板特征参数,根据模板特征参数和TM-align结构比对结果,建立基于特征参数的打分函数Fdscore,并实现α类蛋白质折叠类型自动化分类。使用相同数据集样本,将Fdscore分类方法与TM-score分类方法对比,Fdscore分类方法的平均敏感性、平均特异性、MCC值分别为71.86%、99.49%、0.69,均高于TM-score分类方法相对应结果。上述结果表明该分类方法可用于α类蛋白质折叠类型的自动化分类。

英文摘要：

Protein folding type （PFT） recognition is important in protein structure analysis. PFT classification is the basis of PFT recognition. Here, a PFT template database was built based upon α protein folding types in the ASTRAL-1.65 database. Template feature parameters （TFPs） reflecting PFT topology structures were extracted. A TFP-based scoring function （Fdscore） was proposed on the basis of TFPs and structural alignment results of TM-align. Subsequently, Fdscore was used for automatic classification ofα protein folding types. The proposed Fdscore PFT classification method was compared with the TM-score method using the same data samples. Experimental results showed that the mean sensitivity, the mean specificity, and the MCC val- ue of the Fdscore method were 71.86%, 99.49%, and 0.69, respectively, each of which was higher than that of the TM-score method. The experimental results indicate that the proposed Fdscore method may be used for automatic classification of α protein folding types.