Distinguishing population stratification from genuine allelic effects with Mx: Association of ADH2 with alcohol consumption

Abstract

A universal problem in genetic association studies is to distinguish associations due to genuine effects of the locus under investigation, or linkage disequilibrium with a nearby locus that has a genuine effect, from associations due to population stratification or other artifacts. Fulker et al. (1999) have suggested a test using unselected sib pairs to distinguish these two causes of association. The test is readily implemented within a standard maximum-likelihood framework using the Mx package. The approach is applied to data on ADH2 genotypes and a measure of alcohol consumption from an Australian DZ twin pair sample. Results indicate that the association of the ADH2*2 allele with lower alcohol consumption cannot be explained by simple admixture and that there may be genuine allelic effects of the locus on alcohol consumption. Power calculations are provided to show that these results are plausible for the sample size in this study and consider the effects of genetic architecture and sample structure on required sample sizes for the Fulker et al. test.