In the past decade, genome-wide association studies (GWAS) have successfully identified new genetic variants for a wide variety of phenotypes [1]. However, despite growing sample sizes, the genetic variants discovered by GWAS generally account for only a fraction of the total heritability of a phenotype [2,3]. More than anything, GWAS has shown that many phenotypes, such as height [4], schizophrenia [5] and BMI [6] are highly polygenic and influenced by thousands of genetic variants with small individual effects, requiring very large sample sizes to detect them.
