Fig. 3From: A simulation study investigating power estimates in phenome-wide association studiesBinary Trait Type I Errors. The plot shows the Type I errors for different parameter settings at cases = 200. Each panel represents the different case-control ratio used for the simulation dataset. The Type I error on the y-axis is calculated based on the number of false positive association below significance level of α = 0.00025. The disease penetrance is represented on the x-axis and each colored point represent different MAF used in the simulationsBack to article page