Skip to main content


Fig. 2 | Genome Biology

Fig. 2

From: Exploring the genetic basis of human population differences in DNA methylation and their causal impact on immune gene regulation

Fig. 2

Genetic control of population differences in DNA methylation levels. a Proportions of CpGs and DMS associated to genetic variants identified in the three meQTL studies: merging the two populations (gray shades), mapping in AFB only (red shades) and in EUB only (blue shades). For each mapping, proportions among all 552,141 tested CpG sites and among DMS are indicated in light and dark colors, respectively. ***Fisher’s exact P < 2.2 × 10−16. b Contour plot of meQTL effects on DMS as a function of their difference in derived allelic frequencies (DAF) between populations. For each of the 8459 DMS for which we detected at least one meQTL, we used a kernel density estimation to draw the contour plot of the effect of the derived allele of the meQTL onto methylation (beta, Y axis) according to the ΔDAF (DAFEUB – DAFAFB, X axis). The coefficient and P value of Pearson’s correlation test are displayed. The marginal distribution of the two variables is displayed: top for ΔDAF, and right for beta. c, d Examples of meQTLs detected in this study. Boxplots represent the distribution of β values as a function of genotype, for AFB (red) and EUB (blue) individuals. The minor allele frequency of each meQTL is presented for each population on the top. Gray lines indicate the fitted linear regression model for β value~genotype for each population. e Fold enrichment of meQTLs associated with DMS in GWAS hits. For each of the 17 parental EFO categories, the fold enrichment, the 95% confidence intervals obtained by bootstrap, and the associated P values are shown

Back to article page