invalid p were excluded. In summary, the above observation indicates that f 11 has a higher impact to the power than f 12 & f 22 do in a rare disease model. As the prevalence increases (K = 0.1 and 0.2), the clustering effect gradually disappears. In each R valued bin when K is large (0.1 or 0.2), though the penetrances are sorted in the same order as mentioned above, the dots represent certain continuity instead of clustering. This shows that, in a common disease model, f 11 is not the only or the most effective factor as it is in a rare disease model. Other factors start to interact with each other. Especially when K = 0.2 and R = 1.0 (the fourth bin in the three graphs on the right), the dots appear in clear fan-shaped sectors. This irregularity can be partially explained by the sensitivity to randomness of the model under such setting, i.e., small changes of the parameters can have high impact on the results.
