We assessed the value of ExAC as a reference dataset for clinical sequencing approaches, which typically prioritize or filter potentially deleterious variants based on functional consequence and allele frequency (AF)6. Filtering on ExAC reduced the number of candidate protein-altering variants by 7-fold compared to ESP, and was most powerful when the highest AF in any one population (“popmax”) was used rather than average (“global”) AF (Figure 4a). ESP is not well-powered to filter at 0.1% AF without removing many genuinely rare variants, as AF estimates based on low allele counts are both upward-biased and imprecise (Figure 4b). We thus expect that ExAC will provide a very substantial boost in the power and accuracy of variant filtering in Mendelian disease projects.
