Selected methodological issues in meiotic mapping of obesity genes in humans: issues of power and efficiency

Abstract

This paper focuses on methods for mapping novel obesity genes in humans via meiotic mapping techniques. By novel we mean genes that are as yet unidentified as playing a role in obesity. We begin by presenting a discussion of why we believe it is important to seek out novel obesity genes and, in particular, novel genes of small effect. In light of the arguably Herculean task of finding genes of small effect with conventional gene mapping methods, we discuss alternative methods and procedures that may enhance our ability to map novel obesity genes of small effect. Many of these methods have been discussed previously in the literature and are summarized here. These include reconceptualizing power in the context of genomewide scans, multivariate linkage approaches, the use of phenotypically extreme subjects, and the use of large sibships. These are discussed in the context of linkage studies. Association studies and disequilibrium mapping are also discussed, and again, issues involving the use of extreme phenotypes and multiple testing are included. We also provide a brief discussion of DNA pooling and transmission disequilibrium tests for quantitative traits. Finally, we advocate data pooling techniques (e.g., meta-analysis) to enhance the power and efficiency of the entire field of the genetics of obesity.