Following-Up Genome-Wide Association Study Signals
Lessons Learned From Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium Targeted Sequencing Study
Family studies have for decades documented a critical role of both common and rare inherited variations contributing to interindividual differences in the risks of common heart, lung and blood diseases and their risk factors. Examples include Mendelian forms of disease such as familial hypercholesterolemia caused by low-density lipoprotein receptor mutations and hypertension resulting from epithelial sodium channel subunit mutations. For most conditions, however, the identity of the contributing loci and allelic variants has remained elusive. In the late 1990s, 2 unlikely scientific tools were combined to seed a revolution in gene and locus discovery for many common disease traits. First, the HapMap Project (www.hapmap.org) produced a collection of haplotype-tagging variants that could capture information across the entire genome in common variants potentially influencing risk to disease. Second, large deeply phenotyped cohort studies in both the United States and Europe formed collaborations that took advantage of available DNA, appropriate informed consent, and high-quality longitudinal information about multiple phenotypes.1 These tools formed the basis of many genome-wide association studies (GWAS) which have identified hundreds of loci underlying common chronic conditions, infectious diseases, and cancer susceptibilities (www.genome.gov/gwastudies/).
Because of the small effect sizes expected for most loci of interest, the large number of marker loci measured, and thus the corresponding number of statistical tests, large sample sizes are the norm for most contemporary GWAS. To obtain such sample sizes, multiple groups and studies have combined to form collaborative consortia. For instance, Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE)1 Consortium, which focuses on the genetics of cardiovascular disease and aging, has produced more than 300 publications reporting the results of GWAS for many cardio-, renal-, and cerebrovascular disease-related traits, including metabolic and inflammation traits. Several commentators have raised questions about the utility of GWAS, which have been celebrated …