Undoubtedly, in the near future many more GWA studies with the 500K and even larger SNP arrays will be reported. Such studies will however not cover all genetic variations in the genome [210], because SNPs with a low minor allele frequency (MAF) (<0.05) are usually not included on the arrays, thereby excluding analysis of rare genetic variants. This is unfortunate, since the rare genetic variants are generally considered to have a higher chance of being causative [211]. Thus, candidate gene approaches of selected SNPs with a low prevalence may increase the chances to identify functional genetic variants. Because one would expect that causal SNPs have an effect in any population, a further consideration may involve a choice of SNPs with a low MAF in all ethnical populations. In this connection, one has to be aware of the possibility that such a SNP may need an additional polymorphism(s) to explain the phenotype (multiple genes hypothesis), while the additional genetic variation(s) may not be present in a particular ethnical population. One of the practical problems in dealing with low-MAF SNPs concerns the
