et al., 2006]; matching cases and controls of similar ancestry [Luca et al., 2008]; or analyzing the homogeneous clusters using the Cochran-Mantel-Haenszel test. For most situations these approaches are likely to lead to similar results. Successfully finding the hidden structure using a small number of eigenvectors is the key to controlling for the effects of ancestry. Spectral-GEM should be a useful addition to the tools used to analyze genome-wide association data and other data sets in which ancestry is of interest.
