Comparison of three methods to estimate genetic ancestry and control for stratification in genetic association studies among admixed populations

Abstract

Population stratification may confound the results of genetic association studies among unrelated individuals from admixed populations. Several methods have been proposed to estimate the ancestral information in admixed populations and used to adjust the population stratification in genetic association tests. We evaluate the performances of three different methods: maximum likelihood estimation, ADMIXMAP and Structure through various simulated data sets and real data from Latino subjects participating in a genetic study of asthma. All three methods provide similar information on the accuracy of ancestral estimates and control type I error rate at an approximately similar rate. 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. We demonstrate that approximately 100 ancestry informative markers (AIMs) are required to obtain estimates of ancestry that correlate with correlation coefficients more than 0.9 with the true individual ancestral proportions. In addition, after accounting for the ancestry information in association tests, the excess of type I error rate is controlled at the 5% level when 100 markers are used to estimate ancestry. However, since the effect of admixture on the type I error rate worsens with sample size, the accuracy of ancestry estimates also needs to increase to make the appropriate correction. Using data from the Latino subjects, we also apply these methods to an association study between body mass index and 44 AIMs. These simulations are meant to provide some practical guidelines for investigators conducting association studies in admixed populations.

Analysis of admixture proportions in seven geographical regions of the state of Guerrero, Mexico

“…Ancestry informative markers (AIMs) are genetic loci showing large geographic differentiation that can be used to determine ancestral contributions at the individual and population level (Collins‐Schramm et al, 2004; Tsai et al, 2005). Multiple different panels of AIMs to infer ancestry proportions in recently admixed populations from the Americas have been used (Galanter et al, 2012; Mao et al, 2007; Price et al 2007; Watkins et al, 2012; Yaeger et al, 2008).…”

Evaluación del número mínimo de marcadores para estimar ancestría individual en una muestra de la población argentina

“…Los principales refieren a la cantidad de AIMs empleada, su capacidad informativa, la elección de las poblaciones parentales y el número de individuos que integran las mismas (e.g. Tsai et al 2005). La importancia del número de marcadores utilizados consiste no solamente en su influencia sobre los resultados obtenidos, sino también porque allí radica mayormente el costo que conlleva este tipo de estudios.…”

“…Esto se condice con el hecho de que el único caso para el cual se observó que la varianza entre las cinco corridas supera a la variación entre individuos corresponde al componente subsahariano (Tabla 2). Tsai et al (2005) proponen el uso de al menos 100 marcadores para una estimación correcta. Sin embargo, los autores también indican que el número necesario disminuye a medida que aumenta la capacidad informativa, puesto que al usar AIMs con una diferencia de más de 0,5 entre las frecuencias alélicas de las poblaciones parentales, se requerirían menos de 50 marcadores para mejorar la exactitud de las estimaciones.…”

“…RuizLinares et al 2014;Ruiz-Narváez et al 2010;Tsai et al 2005;Turakulov y Easteal 2003), la gran mayoría utiliza paneles conformados por genotipos basados en grandes cantidades de marcadores en individuos que no provienen de poblaciones argentinas (e.g. HGDP-CEPH - Cann et al 2002;HapMap Project -International HapMap Consortium 2003).…”

African Americans with Asthma: Genetic Insights

“…This information can be used to adjust association test statistics and correct for hidden population stratification in case-control designs using self-reported ethnic membership (92)(93)(94)(95)(96). Simulation studies have demonstrated that upward of 100 AIMs are needed for optimal adjustment (97,98). For example, in a reanalysis of previously published findings based on 35 AIMs (99), Halder and colleagues selected 177 AIMs for maximum information content.…”