Empirical power and total number of false positives generated by each method over 50 simulated data sets simulated under Scenario 2 but with 2000 cases, 2000 controls, and genotype relative risks 1:1.4:1.6
Scenario 2: Larger sample sizes and weaker disease model | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CHISQ | GC | SLR | BLR | STR2 | AM2 | |||||||
| All SNPs included | ||||||||||||
| Power (%) | 88 | 84 | 84 | 88 | 72 | 84 | ||||||
| Total no. false +ve | 4 | 1 | 1 | 1 | 0 | 3 | ||||||
| Causal SNP excluded | ||||||||||||
| Power (%) | 56 | 52 | 52 | 54 | 38 | 56 | ||||||
| Total no. false +ve | 4 | 1 | 1 | 2 | 1 | 2 | ||||||
| Gene 1 excluded | ||||||||||||
| Total no. false +ve | 4 | 1 | 2 | 2 | 1 | 1 | ||||||