单体型组装问题计算模型的比较与分析  

Survey on Computing Models for Haplotype Assembly Problem

作　　者：周伟[1] 王建新[1] 谢民主[1] 陈建二[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083

出　　处：《计算机科学》2008年第11期166-169,共4页Computer Science

基　　金：国家自然科学基金重点项目:生物信息学中的相关组合理论和算法研究(60433020)

摘　　要：单体型检测在遗传病基因的定位、药理反应的研究、个体识别等方面有极其广阔的应用前景。单体型组装问题指如何利用个体的基因测序片断数据,根据不同的优化准则确定该个体单体型的计算问题。对MSR,MFR,MEC,WMLF,MEC/GI等单体型组装模型做了详细的分析比较,得出了如下结论:在没有引入测序误差情况下,上述模型的重构精度基本一致。随着测序误差的增加,MEC/GI模型的容错性最好,重构精度最高;MSR模型受测序误差的影响最大,只适用于测序误差极小的情形。Haplotype detection has expansive application in inherited gene＇s orientation,medicine reaction＇s research and individual identification. The haplotype assembly problem is the computing problem of inducing a individual＇s haplotypes based on several optimal criteria from one＇s DNA fragments sequencing data. This paper made detailed research on such models as MSR, MFR, MEC, WMLF, MEC/GI and got conclusion as follows： in the case of no reading errors, the reconstruction rate of these models is basically identical. As the reading errors increasing, the MEC/GI model suffers reading errors＇ influence least and has the highest reconstruction rate; MSR model suffers reading errors＇ influence most and is only suitable for the case of low reading errors.

关 键 词：单核苷酸多态性 基因型 单体型 NP难问题 

分 类 号：TP241.027[自动化与计算机技术—检测技术与自动化装置] TQ414.1[自动化与计算机技术—控制科学与工程]