中文摘要：

蛋白质折叠规律研究是生命科学领域重要的前沿课题之一,蛋白质折叠类型分类是折叠规律研究的基础。本研究以SCOP数据库的蛋白质折叠类型分类为基础、以Astral SCOPe 2.05数据库中相似性小于40%的α、β、α＋β及α/β类所属的折叠类型为研究对象,完成了989种蛋白质折叠类型的模板构建并形成模板数据库;基于折叠类型设计模板建立了蛋白质折叠类型分类方法,实现了SCOP数据库蛋白质折叠类型的自动化分类。家族模板自洽性检验与独立性检验所得的敏感性、特异性以及MCC的平均值分别为：95.00%、99.99%、0.94与90.00%、99.97%、0.92,折叠类型模板自洽性检验与独立性检验所得的敏感性、特异性以及MCC的平均值分别为：93.71%、99.97%、0.91与86.00%、99.93%、0.87。结果表明：模板设计合理,可有效用于对已知结构的蛋白质进行分类。

英文摘要：

The study of protein folding pattern is one of important topics in life science field. The classification of protein folding type is the base of study on folding pattern. In this paper, protein folding type classification of SCOP database was as the foundation, similarity less than 40% of the folding type was as the object of study in the class of α,β, α＋βand α/β of Astral SCOPe 2.05 database, the template of 989 protein folding types was constructed and the template data base was formed. The method of protein folding type classification was constructed based on the design of folding type template, and the automatic classification of protein folding type of SCOP database was realized. In the self-consistency test and independence test of family template, the mean value of sensitivity, specificity and MCC： 95.00%, 99.99%, 0.94 and 90.00%, 99.97%, 0.92,respectively. In the self-consistency test and independence test of folding type template, the mean value of sensitivity, specificity and MCC： 93.71%, 99.97%, 0.91 and 86.00%, 99.93%, 0.87,respectively. The result shows that the design of template is reasonable and it can be effectively used for the classification of proteins with known structure.