Heritability, Weak Effects, and Rare Variants in Genomewide Association Studies

In less than 5 years, genomewide association studies (GWASs)2 have completely changed the landscape of human genetic research. Our increasing knowledge of the human genome sequence and its variation (http://hapmap.ncbi.nlm.nih.gov/) and technological advances in the design of genotyping microarrays have been instrumental in this evolution; however, an essential factor for success has been the use of large international collaborations for assembling studies that encompass genomewide data for tens or hundreds of thousands of individuals. The number of firmly replicated trait-associated loci that have been identified in GWASs is impressive (http://www.genome.gov/gwastudies), but the contribution of individual single-nucleotide polymorphisms (SNPs) to the studied phenotypes is weak, with rare exceptions. In this article, this aspect of GWASs and some of its implications are discussed. Heritability is the part of the variation of a trait in a population that is contributed by genetic differences among individuals. The heritabilities of common diseases that have been evaluated in family or twin studies are often quite high, frequently reaching 50%. GWAS results provide support for the concept of a polygenic framework underlying complex diseases in which genetic susceptibility (liability) is due to a large number of cumulative weak effects contributed by variants covering the entire range of allele frequencies. According to this model, common disorders may be considered quantitative traits (1).