A model of antigen processing improves prediction of MHC I-presented peptides

Computational prediction of the peptides presented on MHC class I proteins is an important tool for studying T cell immunity. The data available to develop such predictors has expanded with the use of mass spec to identify naturally-presented MHC ligands. In addition to elucidating binding motifs, the identified ligands also reflect the antigen processing steps that occur prior to MHC binding. Here, we developed an integrated predictor of MHC I presentation that combines new models for MHC I binding and antigen processing. Considering only peptides first predicted by the binding model to bind strongly to MHC, the antigen processing model is trained to discriminate published mass spec-identified MHC I ligands from unobserved peptides. The integrated model outperformed the two individual components as well as NetMHCpan 4.0 and MixMHCpred 2.0.2 on held-out mass spec experiments. Our predictors are implemented in the MHCflurry package, version 1.6.0 (github.com/openvax/mhcflurry). See preprint at: https://doi.org/10.1101/2020.03.28.013714 Additional File 1. MULTIALLELIC benchmark dataset with predictions (CSV). Additional File 2. MONOALLELIC benchmark dataset with predictions (CSV). Additional File 3. Training data for MHCflurry 1.6.0 binding affinity (BA) predictors (CSV). Additional File 4. Training data for the variant of MHCflurry BA evaluated on the MONOALLELIC benchmark in Fig. S1 (CSV). Additional File 5. Training data for MHCflurry 1.6.0 antigen processing (AP) predictors (CSV). Additional File 6. Training data for MHCflurry 1.6.0 presentation score (PS) predictors (CSV). Additional File 7. Position weight matrices for the MHCflurry antigen processing predictors (XLSX). Additional File 8. Antigen processing predictions for proteosome-cleaved peptides identified using “mass spectrometry analysis of proteolytic peptides” (MAPP) by Wolf-Levy et al.