Methods
Colocalization and Mendelian randomization analyses were performed. Briefly, colocalization compares the proportionality of the genetic associations for two traits within a given genetic locus, investigating whether the data supports a model with a shared causal variant for both traits3. In Mendelian randomization, genetic variants that proxy the effect of varying the exposure are used to investigate its effect on an outcome4. Further details are provided in the Supplementary Methods.
Summary statistics data for the genetic associations of single-nucleotide polymorphisms with each trait were obtained from publicly available large-scale genome-wide association studies (GWAS) performed on individuals of European ancestries. These data can be obtained from the original studies, as detailed below. Genetic associations with circulating GDF15 levels were based on 3,301 healthy individuals5, genetic associations with BMI were obtained from a GWAS meta-analysis of 806,834 individuals6, and genetic associations with type 2 diabetes liability were based on a GWAS meta-analysis of 74,124 cases and 824,006 controls7. Relevant ethical approval and participant consent were obtained by the original studies.