Although most GWAS of complex traits and diseases have focused on common variants that are typed on genotyping arrays or imputed from population reference panels, rare variant associations may also provide a rich source of biological insights, particularly for traits under strong negative selection37,38. Because association tests of individual rare variants are likely to be underpowered, rare variant association tests generally aggregate evidence for association across multiple rare variants at a locus. In exome sequencing studies (or exome array studies), rare variants are aggregated at the gene level, making the gene the unit of association. This can be done either using burden tests, which assume that all rare variants in a candidate gene have the same direction of effect, or using overdispersion tests, which assume that rare variants in a candidate gene can impact a complex trait in either direction; hybrid omnibus tests are also possible39. Recent studies have shown that both burden tests and overdispersion tests can be performed using only summary association statistics from each rare variant, together with summary LD information40–42 (see Box 2). Roughly, burden tests
