Although our simulations only show that synthetic associations are likely to occur, coupling this demonstration with the available data does suggest that some of the reported associations are likely to be due to this effect, and many more may be enhanced by the signal of surrounding rare causal variants. First, despite considerable efforts, the vast majority of genome-wide associations have never been tracked to causal sites, even though many surrounding regions have been extensively resequenced [2]–[4]. If all of the responsible variants were common SNPs, one might expect that more clear evidence of causation would have been identified by now for a nontrivial number of common variants. Although this expectation is valid for common causal variants, because we know roughly where to look in the genome, this does not hold for synthetic associations due to rare variants, which may reside at a considerable distance from the associated common variants. Second, it is now known that rare variants contribute to common diseases, and that cases that carry the rare high-penetrant contributors to disease often have “typical” clinical presentations [19]–[21]. On balance,
