Linkage mapping of beta 2 EEG waves via non-parametric regression
Linkage mapping of beta 2 EEG waves via non-parametric regression
复制标题
DOI：
10.1002/ajmg.b.10057
复制
发表时间：
2003-04-01
影响因子：
2.8
通讯作者：
Reich, T
中科院分区：
文献类型：
--
作者：
Ghosh, S;Begleiter, H;Reich, T
分析数量性状基因座的参数连锁方法对违反性状分布假设很敏感。非参数方法相对更稳健。在这篇文章中，我们修改了Ghosh和Majumder提出的非参数回归程序[2000：美国遗传学杂志66：1046-1061]，使用COGA项目中产生的全基因组数据绘制β 2 EEG波。在1、4、5和15号染色体上获得了显著的连锁发现，在4号和15号染色体上的多个区域也有发现。我们分析的数据，并没有纳入酗酒作为协变量。我们还测试了基因组区域之间的上位相互作用，表现出显着的连锁与EEG表型，并找到一个区域之间的上位相互作用的证据，每个染色体1和染色体4与染色体15上的一个区域。虽然回归出酗酒的影响并不影响连锁的结果，上位相互作用变得统计学上不显着。(C)2003 Wiley-Liss，Inc.
Parametric linkage methods for analyzing quantitative trait loci are sensitive to violations in trait distributional assumptions. Non-parametric methods are relatively more robust. In this article, we modify the non-parametric regression procedure proposed by Ghosh and Majumder [2000: Am J Hum Genet 66:1046-1061] to map Beta 2 EEG waves using genome-wide data generated in the COGA project. Significant linkage findings are obtained on chromosomes 1, 4, 5, and 15 with findings at multiple regions on chromosomes 4 and 15. We analyze the data both with and without incorporating alcoholism as a covariate. We also test for epistatic interactions between regions of the genome exhibiting significant linkage with the EEG phenotypes and find evidence of epistatic interactions between a region each on chromosome 1 and chromosome 4 with one region on chromosome 15. While regressing out the effect of alcoholism does not affect the linkage findings, the epistatic interactions become statistically insignificant. (C) 2003 Wiley-Liss, Inc.