DnaReSM:一个基于多支持度的DNA重复序列挖掘算法  

DnaReSM:A Multi-Supports-based DNA Repetitive Sequences Mining Algorithm

作　　者：熊赟[1] 陈越[1] 朱扬勇[1] 

机构地区：[1]复旦大学计算机与信息技术系,上海200433

出　　处：《计算机科学》2007年第2期211-212,F0004,共3页Computer Science

基　　金：国家自然科学基金项目(60573093)

摘　　要：DNA序列分析研究是生物信息学的重要内容之一。基因组的基因相关区域和基因外区域中含有大量重复序列,尽管目前大多数重复序列的功能还没能肯定,但它们在遗传分析中已起重要作用。挖掘DNA重复序列成为DNA序列分析的关键。自底向上的挖掘算法中间过程产生很多短的、甚至单字符的模式,使得挖掘效率降低;另一方面,目前序列模式挖掘算法在多序列挖掘中表现出高效性,但由于单支持度定义的局限导致无法在挖掘过程中同时找到单条DNA序列中的重复序列,因此不能很好地适用于DNA重复序列挖掘。本文基于新的多支持度序列模式挖掘框架,提出了一种融合自底向上和自顶向下策略挖掘DNA重复序列的新算法DnaReSM,其结果为生物学相关实验提供基础。实验结果表明,DnaReSM探测算法能有效挖掘DNA重复序列。Research on DNA sequence analysis is one of important subjects in Bioinforrnatics. There exist a great numher of repeats in DNA sequences. Presently, most of them have largely unknown functions, but they have played important roles in genetics. Mining DNA repeats is a promising task. The methods of bottom-up pattern generation in mining sequential pattern, which produce considerable short patterns, will reduce efficiency. Furthermore, present algorithms based on the definition of single support are difficult to find DNA repeats, therefore a novel method which combined bottom-up with top-down named DnaReSM based on multiple supports framework is presented to mine DNA repetitive sequences. Our experimental results demonstrate that DnaReSM is efficient.

关 键 词：数据挖掘 DNA序列 重复序列 序列模式 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]