level of α = 0.05/450000 = 1.1×10−7 in 50000 cases and controls, which reflects the current sample size of some of the larger international GWAS consortia. We investigated the effect of varying the amount of variance the allelic score explained in the biological intermediate (σG 2 = 10%, 5%, 1%, 0.1%), the strength of relationship (i.e. linear regression coefficient) between the intermediate and underlying disease liability (β = 0.1, 0.2 or 0.5), and the prevalence of disease (K = 1%, 5%, 10%, 20%) in both scenarios.
