2022 AACR: Mono-allelic immunopeptidomics data from 109 MHC-I alleles reveals variability in binding preferences and improves neoantigen prediction algorithm

Overview 

This study extends the previously published MHC-I, pan-allelic neoantigen prediction algorithm, SHERPA™, with immunopeptidomics from 84 additional mono-allelic transfected cell lines, totaling data from 109 unique alleles. SHERPA achieves model generalizability and 98% population allelic coverage by integrating nearly 500 additional public immunopeptidomics samples. As a result, SHERPA identifies 1.38-fold more immunogenic epitopes than either NetMHCPan-4.1 and MHCFlurry-2.0 and reveals strong correlations between evolutionary divergence and influential binding pocket positions in the MHC allele.

Download