Improved Prediction of MHC II Antigen Presentation through Integration and Motif Deconvolution of Mass Spectrometry MHC Eluted Ligand Data

BARRA, CAROLINA; PETERS, BJOERN; REYNISSON, BIRKIR; HILDEBRAND, WILLIAM H.; KAABINEJADIAN, SAGHAR; NIELSEN, MORTEN

JOURNAL OF PROTEOME RESEARCH

AMER CHEMICAL SOC

Año: 2020 vol. 19 p. 2304 - 2315

1535-3893

Major histocompatibility complex II (MHC II)molecules play a vital role in the onset and control of cellularimmunity. In a highly selective process, MHC II presents peptidesderived from exogenous antigens on the surface of antigen-presenting cells for T cell scrutiny. Understanding the rules definingthis presentation holds critical insights into the regulation andpotential manipulation of the cellular immune system. Here, weapply the NNAlign_MA machine learning framework to analyze andintegrate large-scale eluted MHC II ligand mass spectrometry (MS)data sets to advance prediction of CD4+ epitopes. NNAlign_MAallows integration of mixed data types, handling ligands withmultiple potential allele annotations, encoding of ligand context,leveraging information between data sets, and has pan-specific powerallowing accurate predictions outside the set of molecules includedin the training data. Applying this framework, we identified accurate binding motifs of more than 50 MHC class II molecules described by MS data, particularly expanding coverage for DP and DQ beyond that obtained using current MS motif deconvolution techniques. Furthermore, in large-scale benchmarking, the final model termed NetMHCIIpan-4.0 demonstrated improved performance beyond current state-of-the-art predictors for ligand and CD4+ T cell epitope prediction. These results suggest that NNAlign_MA and NetMHCIIpan-4.0 are powerful tools for analysis of immunopeptidome MS data, prediction of T cell epitopes, and development of personalized immunotherapies.