In racially admixed populations genetic associations may be confounded by population stratification. To control for population stratification, statistical methods that use marker genotype data to infer population structure have been proposed as an alternative to family-based tests of association. However, there is limited empirical data on how these methods perform in real populations. This application will use well characterized populations of Mexican and Puerto Rican asthmatics, their parents, and control subjects recruited from the same sites to examine the effectiveness of approaches to correct for the effects of population stratification on case-control genetic association studies. This application has three specific aims: 1) To test and compare methods of detecting and correcting for population stratification we will genotype a total of 100 ancestral informative markers (AIMs) for 400 asthma cases and an equal number of control subjects.
These AIMs will then be used with three statistical methods developed to detect and correct for population stratification. The number and characteristics of markers required to correct false positive associations between AIMs, asthma and asthma quantitative traits will be evaluated and compared 2) To compare the power of genomically adjusted case-control studies to the TDT. An allele from each of the 100 AIMS will be considered as a risk factor for a simulated """"""""phenotype"""""""". The association between phenotypes and each AIM will be tested with the TDT and with a case-control analysis after adjustment for stratification to compare the false negative rates for these study designs. 3) To use the results from aim 1 and 2 to define an optimal approach for analysis and interpretation of case-control association studies in these populations and apply this approach to analyze the association between asthma and a series of candidate genes. The results of these studies should provide important insights into the optimal methods to control for population stratification in case-control association studies, thereby facilitating the inclusion of admixed populations in future genetic studies of complex diseases such as asthma.
Showing the most recent 10 out of 97 publications
