Although the real problem is much more complicated than this setting, Equation 1 does provide some useful intuition. Firstly, when the relative effect size is large () the correlation between the marker and causal SNP may only need to be weak (r 2≪0.8) for the association to be detected (the expected test statistic is big). Equally, if the relative effect size is small () then even strong or complete association (0.8<r 2≤1) may not generate sufficient power to reject the null hypothesis of no association.
