The beta-version of GWAPower has been used to support the designs of several GWA studies [9-11]. A GWA study was conducted to search for host genetic variants related to HIV set-point variation. In the first stage of the study, 486 HIV infected Caucasian subjects were genotyped, and the associations between the SNPs and one quantitative phenotype, i.e. the viral load at the set-point, were tested [10]. In the second stage, 2068 subjects were genotyped from the same population, to increase the total sample size to 2554 [11]. Using GWAPower, the statistical powers of the two stages of the studies were calculated (Table 2). The results indicated that assuming H = 7%, a sample size of 486 was adequate to provide 80% statistical power to detect the associated SNP; with the sample size increased to 2554, the smallest H that could be detected was 1.4%. In real data analyses, the largest H was estimated at 9.6%.
