一种处理混合患病兄弟对IBD数据连锁分析的方法  

An approach to linkage analysis for mixed affected sib-pair IBD data

作　　者：崔文泉[1] 许晋[1] 

机构地区：[1]中国科学技术大学统计与金融系,安徽合肥230026

出　　处：《中国科学技术大学学报》2009年第3期254-260,共7页JUSTC

基　　金：国家自然科学基金(10471136;10671189);教育部博士点基金;中国科学院知识创新工程(KJCX3-SYW-S02)资助

摘　　要：患病兄弟对(affected sib-pair,ASP)设计是连锁分析中的重要方法,得到了广泛的研究和应用.然而,实际问题中患病兄弟对中常常会混杂部分半兄弟对,而且由于个人隐私等原因,兄弟对关系的信息往往缺失,我们称这种数据为混合患病兄弟对(mixed affected sib-pair,MASP).对混合患病兄弟对的IBD(identical by descent)数据给出了一种进行连锁分析的方法,构造了可识别的统计模型,证明了用于进行连锁分析的似然比检验统计量的渐近零分布服从自由度为3的卡方分布,同时给出了求解参数估计的EM算法.统计模拟表明,该方法在小样本情形下具有良好表现.Affected sib-pair （ASP） linkage method, which is very popular in linkage analysis and was originally developed for the detection of linkage between a genetic marker and a dichotomous trait in human full sib families, has been extensively investigated and applied. However, among the sampled ASPs, there might be a small proportion of them are actually half sib pairs. Due to privacy, and some other reasons, some of the relationship information of affected sib pairs may be missed. This type of data is called mixed affected sib-pair （MASP）. In this paper, based on the distribution of identical by descent （IBD） sharing, a likelihood-based approach is proposed to do linkage analyzing for this kind of MASP sample. The so-called non-identifiability is overcome. And the EM algorithm, which holds very fast convergence rate in common case, is constructed to estimate the corresponding parameters of concern. For linkage test, under the composite null hypothesis with a nuisance parameter expressing the proportion of half sib-pairs over all MASPs, the large null distribution of the likelihood ratio test （LRT） statistic for the identifiable model is the chi squared distribution with three degrees of freedom. Results of an extensive simulation study demonstrate that the proposed method performs well for small to moderate sample sizes.

关 键 词：连锁分析 可识别性 混合患病兄弟对 似然比检验 EM算法 

分 类 号：Q348[生物学—遗传学]