XM: Association Testing on the X-Chromosome in Case-Control Samples With Related Individuals


  • Contract grant sponsor: National Institutes of Health; Contract grant numbers: K01 CA148958, R01 HG001645, R01 HL085197.

Correspondence to: Timothy Thornton, Department of Biostatistics, University of Washington, Campus Box 357232, Seattle, WA 98195-7232. E-mail: tathornt@u.washington.edu


Genetic variants on the X-chromosome could potentially play an important role in some complex traits. However, development of methods for detecting association with X-linked markers has lagged behind that for autosomal markers. We propose methods for case-control association testing with X-chromosome markers in samples with related individuals. Our method, inline image, appropriately adjusts for both correlation among relatives and male-female allele copy number differences. Features of inline image include: (1) it is applicable to and computationally feasible for completely general combinations of family and case-control designs; (2) it allows for both unaffected controls and controls of unknown phenotype to be included in the same analysis; (3) it can incorporate phenotype information on relatives with missing genotype data; and (4) it adjusts for sex-specific trait prevalence values. We propose two other tests, inline image and inline image, which can also be useful in certain contexts. We derive the best linear unbiased estimator of allele frequency, and its variance, for X-linked markers. In simulation studies with related individuals, we demonstrate the power and validity of the proposed methods. We apply the methods to X-chromosome association analysis of (1) asthma in a Hutterite sample and (2) alcohol dependence in the GAW 14 COGA data. In analysis (1), we demonstrate computational feasibility of inline image and the applicability of our robust variance estimator. In analysis (2), we detect significant association, after Bonferroni correction, between alcohol dependence and single nucleotide polymorphism rs979606 in the monoamine oxidases A gene, where this gene has previously been found to be associated with substance abuse and antisocial behavior.