主成分分析方法在候选基因关联检验中的优效性  

The Superiority of Principal Component Analysis in Candidate Gene Association Test

作　　者：王晓霞[1] 祖培福[2] 

机构地区：[1]齐齐哈尔大学理学院,黑龙江齐齐哈尔161006 [2]牡丹江师范学院理学院,黑龙江牡丹江157011

出　　处：《数学的实践与认识》2015年第15期250-257,共8页Mathematics in Practice and Theory

基　　金：黑龙江省自然科学基金(F201334);黑龙江省级重点创新预研(SY2014006);齐齐哈尔大学青年教师科研启动支持项目(2014k-M17)

摘　　要：对候选基因的关联检验,多标记单倍型方法往往要比单标记方法表达出更多的信息,但是单倍型的数量往往会随着所标记的SNP的数目增多而急剧的增加,这又会大大增加检验统计量的自由度,通过使用统计学中的主成分分析法来降低单倍型空间的维数来检验一个数量性状与多个单倍型的关联情况,并与传统的方法做对比,模拟结果显示,此检验方法有较好的第一类错误率及功效.For the candidate gene association test, It is widely recognized that multi-marker haplotypes are more informative than a single marker. But, the number of haplotypes increase with the increase in the number of typed SNPs, As a result, large numbers of haplotypes will introduce large degrees of freedom in haplotype-based tests.In this article, we propose using the principal component method of statistics to reduce the dimension of tests, and then construct association test to test the association between haplotypes and a quantitative trait, and compared with the traditional ways of test. Our simulation results show that the proposed test has good Type I error rate and Power.

关 键 词：主成分分析 单倍型 功效 关联检验 

分 类 号：Q78[生物学—分子生物学]