模糊聚类分析方法在DNA序列分类中的应用

该文采用模糊聚类分析的方法对DNA序列进行分类.首先从DNA序列中单个碱基分布的“密度”角度出发,提取出DNA序列的特征,然后用模糊聚类分析中常用的方法对DNA序列进行分类.该文运用自行研制开发的集成11种模糊聚类分析算法的模糊聚类分析运算工具,首先对已知的1-20个DNA序列进行模糊聚类分析,根据分类结果的精度,找出了较优的6种聚类分析算法,然后用余下的21-40个DNA序列进行分类;最后,本文一次对所有的1-40个DNA序列进行归类,并综合了所有的分类结果,将难以归类的DNA序列进行了归类.分析结果表明,模糊聚类分析算法具有分类简单且分类结果精度较高的优点.

Application of Fuzzy Cluster in Classification of DNA

This paper presents the method of applying fuzzy cluster in DNA classification.From the density aspect of single base of each DNA's sequence,we distilled its features,and then we used the usual algorithms of fuzzy cluster to analyse the features.We developed a fuzzy cluster calculating tool which is an integration of 11 usual fuzzy cluster algorithms.In this paper,first,we used this tool to analyse the known 1-20 DNA sequences,according to each algorithm's precision,we got six fuzzy clusters having higher precision,and then we used these algorithms to classify the rest 21-40 sequences.At the end of this paper,we classified all the DNA sequences.After synthesizing all the results of each algorithm,we got the classifying result.From the results of analysis,we found the fuzzy cluster in DNA classification was simple and the result had better precision than that of other classifying methods.