A GWAS assessment of the contribution of genomic imprinting to the variation of body mass index in mice.
A GWAS assessment of the contribution of genomic imprinting to the variation of body mass index in mice.
复制标题
DOI：
10.1186/s12864-015-1721-z
复制
发表时间：
2015-08-05
期刊：
影响因子：
4.4
通讯作者：
Gianola D
中科院分区：
文献类型：
--
作者：
Hu Y;Rosa GJ;Gianola D
基因组印记是一种表观遗传机制，它可以导致不同的基因表达，这取决于所接收的等位基因的父母来源。虽然大多数关于印迹的研究都是针对其潜在的分子机制或试图发现可能受印迹影响的基因组区域，但很少有人关注由这种表观遗传过程引起的表型变异的数量。在这篇报告中，我们简要回顾了定量遗传学框架下的单位点印迹模型，并根据该模型提供了遗传方差的分解。从所提出的印迹模型的分析推论表明，印迹对复杂性状的遗传变异有不可忽视的贡献。此外，我们对小鼠身体质量指数（BMI）进行了全基因组扫描分析，旨在揭示在遗传分析中忽略现有印记效应时的潜在后果。使用10021个SNP标记，使用添加剂和印迹模型对小鼠BMI进行全基因组单标记回归。印记显著的标记表明BMI受印记影响。当在分析中考虑印迹效应时，显著方差从1.218 ×10−4变为1.842 ×10−4，这意味着如果不考虑现有的印迹效应，将会丢失三分之一的显著方差。当同时使用标记和家系信息时，考虑印迹的估计遗传率从0.176增加到0.195。当一个复杂的性状受印记影响时，使用忽略这一现象的加性模型可能会导致对加性变异的低估，从而可能导致对该性状潜在遗传结构的错误推断。这可能是解释全基因组关联研究（GWAS）中普遍观察到的部分遗传力缺失的一个可能因素。
Genomic imprinting is an epigenetic mechanism that can lead to differential gene expression depending on the parent-of-origin of a received allele. While most studies on imprinting address its underlying molecular mechanisms or attempt at discovering genomic regions that might be subject to imprinting, few have focused on the amount of phenotypic variation contributed by such epigenetic process. In this report, we give a brief review of a one-locus imprinting model in a quantitative genetics framework, and provide a decomposition of the genetic variance according to this model. Analytical deductions from the proposed imprinting model indicated a non-negligible contribution of imprinting to genetic variation of complex traits. Also, we performed a whole-genome scan analysis on mouse body mass index (BMI) aiming at revealing potential consequences when existing imprinting effects are ignored in genetic analysis. 10,021 SNP markers were used to perform a whole-genome single marker regression on mouse BMI using an additive and an imprinting model. Markers significant for imprinting indicated that BMI is subject to imprinting. Marked variance changed from 1.218 ×10−4 to 1.842 ×10−4 when imprinting was considered in the analysis, implying that one third of marked variance would be lost if existing imprinting effects were not accounted for. When both marker and pedigree information were used, estimated heritability increased from 0.176 to 0.195 when imprinting was considered. When a complex trait is subject to imprinting, using an additive model that ignores this phenomenon may result in an underestimate of additive variability, potentially leading to wrong inferences about the underlying genetic architecture of that trait. This could be a possible factor explaining part of the missing heritability commonly observed in genome-wide association studies (GWAS).