基于蛋白质二级序列的关联多分类算法  

Association multi-classification algorithm based on protein secondary structure sequence

作　　者：杨炳儒[1] 周谆[1] 侯伟[1] 

机构地区：[1]北京科技大学信息工程学院,北京100083

出　　处：《系统工程与电子技术》2010年第6期1318-1324,共7页Systems Engineering and Electronics

基　　金：国家自然科学基金重点项目(60675030);北京市自然科学基金(4022008)资助课题

摘　　要：蛋白质二级结构预测是公认的生物信息学领域的国际性难题。以基于内在认知机理的知识发现理论(knowledge discovery theory based on inner cognitive mechanism,KDTICM)理论的扩展性研究与数据库中的知识发现(knowledge discovery in database*,KDD*)模型为基础,提出一种基于结构序列的多分类算法——SAC(structuralassociation classification),可以有效地解决蛋白质二级结构预测问题。该算法借助设定支持度阈值的精化知识库的方法,其预测准确率能够超过85%。以该算法为核心,构建了一个蛋白质二级预测模型——复合金字塔模型。实验证明,在RS126、CB513I、LP数据集上的预测准确率均超过80%,超过目前已知的国际主流水平。The prediction of protein secondary structure is one of the major issues in Bioinformatics.As one of the researches of KDTICM theory,a multi-classification algorithm based on structure sequence is proposed,which is based on knowledge discovery in database＊（KDD＊） model.The SAC algorithm can effectively solve the problem of protein secondary structure prediction.The algorithm＇s accuracy exceeded by 85% by using the reduction of knowledge base through the setting of the confidence threshold value.A compound pyramid model is built with the SAC algorithm being regarded as a kernel.Experimental results show that the predictive accuracy exceeded by 80% when using in the datasets of RS126,CB513 and ILP,which is equivalent or even excels known national and international levels.

关 键 词：关联分类 蛋白质二级结构预测 数据库中的知识发现 复合金字塔模型 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]