Indeed, functional classification of sequence variations that have been identified by GWA studies to be associated with complex traits, albeit still strongly focused on those of medical importance, shows that the vast majority of variations are located in noncoding regions [104],[105]. In most cases, the causative mutations have yet to be defined, and their mechanistic basis is unknown—especially whether they affect cis-acting binding sites for regulatory proteins or the function or expression of regulatory RNAs. One good candidate for the latter is the ncRNA ANRIL, which lies antisense to CDKN2A and traverses a noncoding region centromeric to CDKN2A, a region implicated in a range of complex diseases including cancer, type 2 diabetes, periodontitis, and coronary heart disease [106]–[112]. Perplexingly, however, those variants mapped by GWA studies to date account for only a small proportion of genetic variation in disease or quantitative traits [113]. These traits are clearly multifactorial and may be affected by rare variants with strong effects that have yet to be recognized.
