Fig. 1: Manhattan and qq-plots of the associations from the epigenome-wide association meta-analysis on DKD in type 1 diabetes using three different analysis models (a, b and c). | Nature Communications

Fig. 1: Manhattan and qq-plots of the associations from the epigenome-wide association meta-analysis on DKD in type 1 diabetes using three different analysis models (a, b and c).

From: Epigenome-wide meta-analysis identifies DNA methylation biomarkers associated with diabetic kidney disease

Fig. 1: Manhattan and qq-plots of the associations from the epigenome-wide association meta-analysis on DKD in type 1 diabetes using three different analysis models (a, b and c).The alternative text for this image may have been generated using AI.

The models were adjusted for (a) age, sex and six WCCs (Minimal model), (b) Minimal model and current smoking status, and (c) minimal model and current smoking status, HbA1c, high-density lipoprotein cholesterol, triglycerides, duration of diabetes and body mass index. The x-axis shows the chromosomal locations, and the y-axis shows −log10(p-values). P-values in the EWAS (FinnDiane and UK-ROI) were derived from hierarchal linear models and FDR-adjusted, after which they were meta-analysed using METAL. The red line indicates the epigenome-wide significance threshold (p ≤ 9.9 × 10−8). Gene symbols (or CpG identifier if intergenic) are displayed for differentially methylated CpGs that reached epigenome-wide significance across all models.

Back to article page