Their prostate cancer study had a total of 2277 subjects, approximately half of which were cases, again split in a 9∶1 ratio and a marker panel of 165,508 nearly independent SNPs. Assuming a prevalence of 2.4% and sibling relative risk of 2.8 [29], this design has 19% power if the markers explain the full heritability. If the sample were split in a 1∶1 ratio, the power would be 42%. It is clear that even with the optimistic assumption that the markers explain the full heritability, this study was unlikely to detect an association of the polygenic score for either cancer.
