Figure 2

Standardized β-estimates from linear regression analysis with the amount of visceral (VAT; left panel) or subcutaneous (SAT; right panel) adipose tissue as exposure and urine metabolites as outcome conducting either the whole population (square), only men (circle) or women (diamond). Displayed are only metabolites which were annotated and significant (controlling the false discovery rate (FDR) at 5%) in at least one of the subsets (indicated by darker colors). Metabolites printed in bold showed a nominal significant (p < 0.05) interaction term between VAT or SAT and sex. Regression models were adjusted for age, (sex), smoking behavior, alcohol consumption, LDL-cholesterol, systolic blood pressure, and estimated glomerular filtration rate. The Venn diagram displays the overlap in associated metabolites, including unknown(*) compounds. PDG = 5beta-pregnan-3alpha,21-diol-11,20-dione 21-glucosiduronate.