engleski

Additional variance of serum lipid levels explained by incorporating less significant genetic variants and allelic heterogeneity

Serum concentrations of total cholesterol, low- density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides are important heritable risk factors for cardiovascular disease. Large-scale genome-wide association studies (GWAS) have identified many genetic variants robustly associated with serum lipid levels ; however, these variants explain only a small proportion of the overall phenotypic variance, leading to the issues of `missing' heritability. In this study we examined two possible sources of missing heritability: first, variants with smaller effects whose associations with the trait failed to reach genome-wide significance and second, allelic heterogeneity due to the effects of multiple variants at a single significant locus. For this purpose, we developed an analytical approach that accounts for local linkage disequilibrium patterns and dissects the allelic heterogeneity of significant loci from summary-level statistics from a meta-analysis of GWAS. We applied this method to the summary results reported by Teslovich et al, Nature (2010) and dissected heterogeneous allelic effects of lipids associated loci identified at various significance levels. In a sample of 1, 304 individuals from an island population of the Adriatic coast of Croatia, we further examined the extent of phenotypic variance of serum lipid levels explained by incorporating the effects of both the leading and additional variants of lipids-associated loci. Our results indicate that approximately half of the lipid loci that achieved stringent genome-wide significance (p-value < 5×10−8) showed significant evidence of allelic heterogeneity. In addition, including less significant loci (p-value < 5×10−5) and accounting for effects of allelic heterogeneity substantially improved the variance explained of serum lipid levels in our samples (>50% higher than the variance explained by using only leading SNPs with genome-wide significance). These observations suggest that allelic heterogeneity is common phenomenon in complex traits and an appreciable fraction of SNPs not meeting genome-wide significance might have small but genuine effects.

serum lipid levels; cardiovascular disease; allelic heterogeneity

nije evidentirano