基于最小生成树的多序列联配算法  

Multiple Sequence Alignment Based Minimum Spanning Tree

作　　者：胡桂武[1] 郑启伦[1] 彭宏[1] 邓伟林[2] 

机构地区：[1]华南理工大学计算机科学与工程学院 [2]广东职业技术学院计算机系,广州510300

出　　处：《计算机科学》2005年第4期59-61,共3页Computer Science

基　　金：国家自然科学基金(编号:30230350)

摘　　要：多序列联配(MAS)是现代生物信息学中的重要工具之一,MAS问题是NP-难的,因此需要一些启发式方法在合理的时间内联配大的数据集。本文提出了一个基于最小生成树的多序列联配算法,并使用BALiBASE标准数据集合,对我们的算法进行了性能评价,结果表明算法较之ClustalX类的算法其精确度更高。Multiple Sequence Alignment (MSA) is one of the most important tools in modern biology. The MSA problem is NP-hard; therefore, heuristic approaches are needed to align a large set of data within a reasonable time. In this paper, a new MSA algorithm is proposed. We use a Minimum Spanning Tree (MST) algorithm to construct a guide tree in which the sequences are aligned. Quality assessment of our algorithm with ClustalX was conducted using the BAliBASE benchmarks. It is found that our algorithm can provide alignments which are better than those from ClustalX in most test cases.

关 键 词：最小生成树 多序列联配算法 生物信息学 数据集 数据库 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]