Logistic regression protects against population structure in genetic association studies

Table 7.

Empirical power and average number of false positives generated by each method in 50 simulated data sets simulated under Scenario 3 but with cases chosen only from one population


Scenario 3: All cases from one population

CHISQ
GC
SLR
BLR
STR2
AM2
All SNPs included
   Power (%) 98 0 86 92 92 98
   Average no. false +ve 52 0 6 13 7 29
Causal SNP excluded
   Power (%) 98 0 68 80 78 94
   Average no. false +ve 52 0 6 14 7 31
Gene 1 excluded
   Average no. false +ve
52
0
6
15
6
30

This Article

  1. Genome Res. 16: 290-296

Preprint Server