不加权算术平均组对方法的改进及应用  

Improvement for Unweighted Pair Group Method With Arithmetic Mean and Its Application

作　　者：李玉鑑[1] 徐立业[1] 

机构地区：[1]北京工业大学 计算机学院,北京100022

出　　处：《北京工业大学学报》2007年第12期1333-1339,共7页Journal of Beijing University of Technology

基　　金：北京市自然科学基金(4052005);北京市属市管高等学校'中青年骨干教师培养计划'项目PHR(IHLB).

摘　　要：为了解决传统不加权算术平均组对方法(unweighted pair group method with arithmetic mean,简称UPG- MA)存在的'tie trees'问题,通过改进UPGMA,提出了不加权算术平均组群方法(unweighted multiple group method with arithmetic mean,简称UMGMA),从理论和应用上证明了UMGMA能产生唯一的进化树,并且在UPGMA树唯一时,UMGMA树和UPGMA树在不计分支次序时完全相同,解决了UPGMA树的唯一性问题.与UPGMA不同之处在于,UMGMA反复利用极大紧邻子树上的顶点把多个距离最近的种群进行合并,因此在UPGMA产生的二叉树不唯一时,UMGMA能产生一棵具有唯一拓扑结构的多叉树.通过适当选择大于0的容差参数,UMGMA还可以在不同的宏观层次上产生容差进化树,以突出物种较多时进化树的整体脉络.The problem of ＂tie trees＂ has been noticed in the traditional Unweighted Pair Group Method with Arithmetic Mean （UPGMA）. In order to solve the problem, this paper presents a method of Unweighted Multiple Group Method with Arithmetic Mean （UMGMA）, which is an improvement for UPMGA. It has been shown in theory and application that UMGMA can always produce a unique phylogenetic tree. In the case when the UMGMA tree is bifurcating, it must be the same as the corresponding UPGMA tree without considering the order of branches. Differing from UPGMA, UMGMA repeatedly merges multiple groups into one by the vertices of a maximal θ-distant subtree until only one group remains, thus it may produce a multifurcating tree with a unique topology. By choosing a tolerant parameter properly, UMGMA can be used to construct several unique tolerant multifurcating trees that outline the major branches of a phylogenetic tree on different levels.

关 键 词：数据处理 不加权算术平均组对方法 系统发育分析 二叉树 多叉树 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]