Genome-wide association mapping has become widely popular in recent years as a method to dissect the genetic architecture of traits. With the $1,000 dollar genome on the horizon, genome-wide association studies are being conducted in large samples, often combining several human sub-populations in meta studies. These data are challenging to analyze as not only are the samples typically structured but most traits of interest (such as human diseases) are usually also complex and polygenic. ❧ For complex traits, linkage disequilibrium can cause misleading or synthetic associations in genome-wide association studies. On a local scale, synthetic associations caused by linkage have received increased attention recently, highlighting the difficulties involved with fine-mapping traits. Linkage disequilibrium on a global scale, e.g. caused by population structure, is similarly also a source for confounding. Mixed linear models have been shown to handle the confounding effects of a diffuse background of large numbers of loci of small effect well, but will not generally be able to account for multiple causative loci of moderate effect. ❧ This dissertation addresses common problems encountered when mapping complex traits in structured populations and describes how linear mixed models can be used to tackle these. Chapter 2 presents the results from a genome-wide association study of 107 traits in very structured set of 199 Arabidopsis thaliana accessions, where I show how most of the 107 traits are significantly confounded by population structure as well as discuss a handful of interesting associations, both simple and complex. In chapter 3 a multiple loci mixed model is proposed as a general method for mapping complex traits in structured populations. By including markers as cofactors in the model the linear mixed model is dynamically updated, resulting in improved power to detect associations in simulated data (when restricting to small false discovery rates). Finally, in chapter 4 I present an approach for combining correlated traits in one linear mixed model which allows for analysis of pleiotropic effects in complex traits confounded by population structure. Using both human and A. thaliana examples, interesting associations which were missed in the single trait and single marker analysis, will be identified and discussed.
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.