Figure 3From: Trick or treat: The effect of placebo on the power of pharmacogenetic association studiesThe effect of strategy for selecting markers ( M , at random or categorised) on the power of detecting an association between the markers and the response to the drug. The results are based on 100 simulations. Estimated power is shown for the causative single nucleotide polymorphism (SNP) (-x-), randomly selected SNPs (-■-) and a categories-based strategy (-▲-). (a) M = 3, genotype relative risk (GRR) = 3, f2 = 0.6, f0 = 0.2; (b) M = 5, GRR = 3, f2 = 0.6, f0 = 0.2; (c) M = 5, GRR = 3, f2 = 0.8, f0 = 0.266; (d) M = 5, GRR = 4, f2 = 0.8, f0 = 0.2; (e) M = 3, GRR = 4, f2 = 0.8, f0 = 0.2; (f) M = 5, GRR = 4, f2 = 0.8, f0 = 0.2.Back to article page