Estimation of admixture and detection of linkage in admixed populations bya Bayesian approach: application to African-American populations

We describe a novel method for analysis of marker genotype data from admire d populations, based on a hybrid of Bayesian and frequentist approaches in which the posterior distribution is generated by Markov chain simulation an d score tests: are obtained from the missing-data likelihood. We analysed d ata on unrelated individuals from eight African-American populations, genot yped at tell marker loci of which two (FY and AT3) are linked (22 cM apart) . Linkage between these two loci was detected by testing for association of ancestry conditional on parental admixture. The strength of this associati on was consistent with European gene flow into the African-American populat ion between five and nine generations ago. To mimic the mapping of an unkno wn gene in an 'affecteds-only' analysis, a binary trait was constructed fro m the genotype at the AT3 locus and a score test was shown to detect linkag e of this 'trait' with the FY locus. Mis-specification of the ancestry-spec ific allele frequencies - the probabilities of each allelic state given the ancestry of the allele - was detected at three of the ten marker loci. The methods described here have wide application to the analysis of data from admired populations, allowing the effects of linkage and population structu re (variation of admixture between individuals) to be distinguished. With m ore markers and a more complex statistical model, genes underlying ethnic d ifferences in disease risk could be mapped by this approach.