Fig. 4: Causal networks between the 5 MetBAGs and 525 Des.

We constructed a causal network by employing bi-directional two-sample Mendelian randomization, following a rigorous quality control procedure to select exposure and instrumental variables (number of IVs > 7), corrected for multiple comparisons (based on the number of DEs), and performed sensitivity analyses (e.g., horizontal pleiotropy) to scrutinize the robustness of our results. Two causal networks were analyzed: (i) MetBAG2DE and (ii) DE2MetBAG. The arrows indicate the direction of the established causal relationship from the exposure variable to the outcome variable. An interactive network visualization is also available at https://labs-laboratory.com/medicine/metbag_mr. It is crucial to approach the interpretation of these potential causal relationships with caution, despite our thorough efforts in conducting multiple sensitivity checks to assess any potential violations of underlying assumptions. DE: disease endpoint; MetBAG: metabolome-based biological age gap.