Fig. 1

Schematic representation of assumptions of our MR analysis. In this investigation, MR has been used to test the causal relationship between BMI and heart attacks in adulthood. The numbers indicate the three assumptions: 1) instrumental SNPs are associated with BMI; 2) instrumental SNPs are not associated with confounders; 3) instrumental SNPs do not have direct effects on the risk of heart attacks, i.e., their effects on heart attacks are mediated only through BMI