Much progress has been made in our understanding of the genetic basis of AD. About 50-60% of the phenotypic variance of AD can be accounted for by genetic factors [8,24]. In recent years, millions of SNPs have been genotyped in genome-wide association studies (GWAS) to identify loci that underlie AD [6,3]. Because a million or more statistical tests are performed in GWAS, a stringent statistical significance threshold (e.g., 5 × 10−8) must be used in GWAS to control the false positive rate. Despite this stringent criterion, several loci have been implicated as contributing to the etiology of AD. To name a few, two closely linked intergenic SNPs (rs7590720, P = 9.72×10−9 and rs1344694, P = 1.69 × 10−8) located on chromosomal region 2q35 were genome-wide significant in a German population [28]. From a GWAS in Japanese, a cluster of 12 SNPs in the ALDH2 gene were significantly associated with AD [25]. The strong effect of ALDH2 on AD risk in Asian populations has been confirmed by meta-analysis [14] and by a recent GWAS conducted by us in a Chinese population [22].
