Comparison of Three Statistical Methods To Estimate Genetic Ancestry and Control for Stratification in Genetic Association Studies among Admixed Populations

• Activity Number: 237
• Type: Contributed
• Date/Time: Tuesday, August 9, 2005 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Hui-Ju Tsai*+
• Companies: University of California, San Francisco
• Keywords: population stratification ; genetic association study ; admixed population ; ancestry estimate

Abstract

Population stratification may confound the results of genetic association studies. Several methods have been proposed to estimate ancestral information in admixed populations and adjust for population stratification. We evaluate the performance of three methods: maximum likelihood estimates, ADMIXMAP, and structure through various simulations and real data from an asthma study. Our results show the most important factor in determining accuracy of the ancestry estimate and in minimizing type-I error rate is the number of markers used to estimate ancestry. For markers of mean informativeness 0.4, approximately 100 markers are required to obtain estimates of ancestry. After accounting for ancestry information, the excess of type-I error rate is controlled at the 5% level when using 100 markers to estimate ancestry. Increasing the number of individuals from ancestral populations has little effect on the accuracy of estimating ancestry. We also apply these methods to test the association between body mass index and 44 ancestry informative markers in real data. The results provide practical guidelines for investigators conducting association studies in admixed populations.