It is becoming clear that many loci harbouring common alleles affecting common diseases will have effect sizes in the range 1.1–1.2, and our simulations demonstrate that there is almost no power to detect these in studies of the size currently underway. As has already been shown empirically [19],[20] these loci can be found by meta-analyses and follow up in larger samples of GWA findings. Slightly larger relative risks do become detectable in large samples. For example the power to detect an effect of size 1.3 jumps from almost zero with 1000 cases and 1000 controls to over 50% in a study three times the size.
