Genome-wide association study and conditional analysis reveal the importance of non-additive effects and ethnicity interaction for coronary heart disease

Abstract

Coronary heart disease (CHD) is a complex genetic etiology, and its incidence is also affected by life styles, such as smoke and physical activities. Genome-wide association studies have identified multiple risk loci associated with CHD. However, genetic risk variants reported to date account for only a small fraction of heritability. Besides, the mechanism of how life styles affect CHD progression still remains vague. In this study, we aimed to explore the missing heritability via introducing non-additive effects into genetic analysis model and to investigate the impacts of life styles on genetic architectures among different ethnic populations. Mixed linear model (MLM) was conducted to identify causal single nucleotide polymorphisms (SNPs) associated with CHD using data from the Multi-Ethnic Study of Atherosclerosis (MESA). Saturated model including genetic effects of additive, dominance, epistasis and gene-ethnicity interaction was adopted to fit the complex genetic architecture of CHD. Each of the six life styles (walk, read, transportation, exercises, TV, and smoke) was set as a cofactor to explore the change of genetic architecture after removing their influences. To facilitate personalized medicine, we also predicted genotypic effects for each locus. There were 61 quantitative trait SNPs (QTSs) and 23 pairs of epistasis detected significantly (P < 0.05). The heritability explainedfrom 64.58%to 74.94%across different models.We observed that additive effects (including both general and ethnic-specific additive effects) contributed only a small portion of heritability, ranging from 3.45% to 5.72%. In contrast, non-additive effects dominated large part of total heritability. Genetic effects attributed to life styles were analyzed by conditional analysis. The conditional analysis demonstrated that life styles exhibited significant impacts on the genetic architecture. In the meanwhile, four ethnic groups exhibited notably distinctive genetic patterns under seven models, indicating genetic heterogeneity for CHD among four races.