Skip to main content
. Author manuscript; available in PMC: 2011 Nov 30.
Published in final edited form as: Ann Appl Stat. 2011 Sep 1;5(3):2052–2077. doi: 10.1214/11-AOAS469

Fig. 2.

Fig. 2

Power comparison for BEAM1, BEAM2, Block1, Block2, and Block3 using simulated data from the HapMap data. Under each simulation setting and from 50 data sets, power (y-axis) is calculated as the proportion of disease-associated SNPs (within 5 SNPs of true disease loci) among top m SNPs (x-axis), ranked by the posterior probability of association. Each data set contains 1,000 candidate SNPs in 1,000 cases and 1,000 controls. The disease allele frequency is 0.05, 0.10, 0.20, and 0.50, respectively. The marginal effect size of each disease SNP (unobserved) is 0.5. Pink: BEAM1; Red: BEAM2; Blue: Block1; Green: Block2; Black: Block3.