Fig. 2From: Assessing causal links between age at menarche and adolescent mental health: a Mendelian randomisation studyDirected acyclic graph illustrating the Mendelian randomisation design. Gj is the jth genetic variant, with direct effect γj on exposure X, and direct effect αj on outcome Y; θ is the estimated causal effect of the exposure on the outcome; ϕj is the relationship between confounders U and Gj; dotted lines represent possible violations of the MR assumptionsBack to article page