We should acknowledge that the distribution of effect sizes of true associations is unknown, and there is no guarantee that they would be similar for different traits. The difficulty of arriving at the true causal variants (which may have larger effect sizes than their markers) adds another layer of complexity. Moreover, given simple power considerations, it is expected that a large proportion of the large effects have been identified, while only a small proportion of the smaller effects and a negligible proportion of the tiny and infinitesimal effects are already discovered. With these caveats, most evidence from GWA studies to-date is more compatible with the scenarios of ORav being in the range of 1.15, [15] but the 1.02 scenario is not implausible, and for some traits the 1.5 scenario may be operating, but we still have not identified the true variants.
