was entirely driven by low frequency SNPs, most notably SNPs with MAF≤2%. More specifically, imputation quality based on the most diverse ALL panel was reduced by an abundance of population-specific SNPs that are likely absent in African Americans but present in other populations with differing MAF and linkage disequilibrium structures. The ALL reference panel had the highest overall imputation quality when these low frequency, population-specific SNPs were not considered, and this panel resulted in the highest quality for common SNPs (MAF>2%), regardless of the population specificity.
