基于最大频繁序列的蛋白质分类算法  

Protein Sequence Classification Algorithm Based on Maximal Frequent Sequence

作　　者：卫锦花[1] 吴陈[1] 

机构地区：[1]江苏科技大学电子信息学院,江苏镇江212003

出　　处：《江苏科技大学学报（自然科学版）》2007年第B12期79-83,共5页Journal of Jiangsu University of Science and Technology:Natural Science Edition

摘　　要：针对现有基于频繁模式的分类算法未考虑完全频繁模式所产生的大量无效序列,提出了一种基于最大频繁序列的蛋白质分类算法,此算法每一类都以独有的最大频繁模式作为代表,执行模式裁减和测试数据分类。实验表明该算法在继承传统算法优点的同时提高了结果的精确度,降低了模式的冗余度,此应用增加了分类的生物信息学意义。Aimed at the massive invalid sequences caused by the complete frequent patterns,which is not considered in the existing classification algorithm, a protein sequence classification algorithm is proposed based on the maximal frequent sequence. In this algorithm each class can be presented by the particular maximal frequent pattern, then the pattern can be reduced and the test data can be classified. Experiments show that this algorithm can improve the precision of results and reduce the redundancy of the pattern with remaining the advantages of the traditional algorithm, the bioinformatics meaning can then be increased through such an application.

关 键 词：蛋白质序列 分类 最大频繁序列 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]