Accurate eplet identification is necessary for accurate risk assessment

Abstract

Krummey et al base their critical opinion on the fact that, in >85% of cases in studies performed by Senev et al and Engen et al, the difference between imputed and sequenced HLA allele assignment was ≤2 eplet mismatches. Their argument is that "if imputation versus high-resolution genotyping increased the mismatch burden by four HLA-DR eplets, but did not change the immunologic risk stratification because the number of eplet mismatches remains below the low-risk cut-off of ten, the imputation would be useful".