We have also made no assumption about which SNPs are more likely to be associated. One commonly held view is that non–synonymous or other coding SNPs are more likely to be associated than random SNPs, but this view is by no means universal, and the evidence is only now accruing through the first generation of scans. Furthermore, in a linkage disequilibrium scan of tag SNPs, it is harder to distinguish functional from non–functional SNPs. Another view is that SNPs in longer blocks of LD are more likely to be associated [Pe'er et al., 2006]. However, in a scan of randomly chosen SNPs, such as those considered here, this prior is likely to be attenuated by the fact that longer blocks of LD are likely to contain more genotyped SNPs [I. Pe'er, Personal Communication]
