variants that have been identified only account for a very small fraction of the heritability. This has been called the “missing heritability problem” (Manolio et al., 2009), and several potential explanations have been put forth for this phenomenon, including the potential importance of rare variants and/or structural variants (genomic changes such as insertions, deletions, inversions, and translocations of stretches of genetic material), neither of which are well captured by current GWAS platforms. In addition, there is the possibility of gene–gene and/or gene–environment interactions. The very small effect sizes associated with identified variants have led to the need for increasingly large consortia and meta-analyses to have reasonable power to detect these small effects; for example, a recent GWAS of body mass index analyzed data from nearly 250,000 individuals (Speliotes et al., 2010)! As studies are combined for these large meta-analyses, the ability to analyze more refined phenotypes is usually compromised, as different assessments often have been used across different studies. One can imagine that the additional layer of trying to find common environmental measures that have been assessed across studies to incorporate gene–environment interaction into these efforts becomes formidable. I often hear social scientists chiding geneticists for “ignoring” the environment; hopefully
