Genome-Wide Association Studies and the Problem of Relatedness Among Advanced Intercross Lines and Other Highly Recombinant Populations

Model organisms offer many advantages for the genetic analysis of complex traits. However, identification of specific genes is often hampered by a lack of recombination between the genomes of inbred progenitors. Recently, genome-wide association studies (GWAS) in humans have offered gene-level mapping resolution that is possible because of the large number of accumulated recombinations among unrelated human subjects. To obtain analogous improvements in mapping resolution in mice, we used a 34th generation advanced intercross line (AIL) derived from two inbred strains (SM/J and LG/J). We used simulations to show that familial relationships among subjects must be accounted for when analyzing these data; we then used a mixed model that included polygenic effects to address this problem in our own analysis. Using a combination of F₂ and AIL mice derived from the same inbred progenitors, we identified genome-wide significant, subcentimorgan loci that were associated with methamphetamine sensitivity, (e.g., chromosome 18; LOD = 10.5) and non-drug-induced locomotor activity (e.g., chromosome 8; LOD = 18.9). The 2-LOD support interval for the former locus contains no known genes while the latter contains only one gene (Csmd1). This approach is broadly applicable in terms of phenotypes and model organisms and allows GWAS to be performed in multigenerational crosses between and among inbred strains where familial relatedness is often unavoidable.SUSCEPTIBILITY to diseases such as drug abuse is partially determined by genetic factors. The identification of the alleles that underlie disease susceptibility is an immensely important goal that promises to revolutionize both the diagnosis and the treatment of human disease. Genome-wide association studies (GWAS) in humans can locate common alleles with great precision. However, GWAS may be unable to identify the bulk of the heritable variability for common genetic diseases; some of this “missing heritability” is thought to be due to rare alleles (Manolio et al. 2009). Model organisms are complementary to human genetic studies and offer unique advantages including the ability to control the environment, perform dangerous or invasive procedures, and test hypotheses by manipulating genes via genetic engineering; a final advantage is that crosses between two inbred strains avoid many of the difficulties associated with rare alleles.Studies in model organisms have frequently employed intercrosses (F₂''s) to identify quantitative trait loci (QTL) that underlie phenotypic variability. F₂ crosses are easy to produce and easy to analyze; however, due to a lack of recombination they can identify only larger genomic regions and are thus unsuitable for identifying the genes that cause QTL (Flint et al. 2005; Peters et al. 2007). This is a serious limitation that can be addressed by using populations with greater numbers of accumulated recombinations. Darvasi and Soller (1995) suggested the creation of advanced intercross lines (AILs) by successive generations of random mating after the F₂ generation to produce additional recombinations. An AIL offers vastly improved mapping resolution while maintaining the desirable property that all polymorphic alleles are common.We used an AIL to study sensitivity to methamphetamine, which is a genetically complex trait that may be useful for identifying genetic factors influencing the subjectively euphoric response to stimulant drugs and susceptibility to drug abuse (Palmer et al. 2005; Phillips et al. 2008; Bryant et al. 2009). For example, a prior study suggested that the gene Casein Kinase 1 Epsilon (Csnk1e) might influence sensitivity to the acute locomotor response to methamphetamine in mice (Palmer et al. 2005). This conclusion has been bolstered by additional pharmacological (Bryant et al. 2009) and genetic studies. In addition, we have shown that polymorphisms in this gene are associated with sensitivity to the euphoric effects of amphetamine in humans (Veenstra-Vanderweele et al. 2006). Another group has subsequently reported that this same gene is associated with heroine addiction (Levran et al. 2008). Thus, genes that modulate the acute locomotor response to a drug in mice may also be important for sensitivity to similar drugs in humans as well as the risk for developing drug abuse.The purpose of this study was to develop a framework for rapid identification of high precision QTL and ideally specific genes that influence sensitivity to methamphetamine in mice by employing an AIL. We produced an F₂ cross (n = 490) and a corresponding 34th generation AIL (n = 688) derived from the inbred strains SM/J and LG/J. This allowed us to compare and integrate the results from F₂ and AIL mice. We examined the locomotor stimulant response to a 2-mg/kg dose of methamphetamine, which is extremely disparate in the two progenitor strains. We performed a GWAS using either simple regression, which ignored relatedness, or a mixed model that accounted for relatedness by using identity coefficients that were calculated from the pedigree. We also explored two methods to estimate significance: simple permutation and gene dropping. We discuss the performance of a mixed model that includes polygenic effects vs. simple regression and the performance of permutation vs. gene dropping. The methods used in this study are applicable to a variety of other phenotypes and populations.