EPIP: MHC-I epitope prediction integrating mass spectrometry derived motifs and tissue-specific expression profiles

Abstract Background Accurate prediction of epitopes presented by human leukocyte antigen (HLA) is crucial for personalized cancer immunotherapies targeting T cell epitopes. Mass spectrometry (MS) profiling of eluted HLA ligands, which provides high-throughput measurements of HLA associated peptides in vivo , can be used to faithfully model the presentation of epitopes on the cell surface. In addition, gene expression profiles measured by RNA-seq data in a specific cell/tissue type can significantly improve the performance of epitope presentation prediction. However, although large amount of high-quality MS data of HLA-bound peptides is being generated in recent years, few provide matching RNA-seq data, which makes incorporating gene expression into epitope prediction difficult. Methods We collected publicly available HLA peptidome and matching RNA-seq data of 34 cell lines derived from various sources. We built position score specific matrixes (PSSMs) for 21 HLA-I alleles based on these MS data, then used logistic regression (LR) to model the relationship among PSSM score, gene expression and peptide length to predict whether a peptide could be presented in each of the cell line. We further built a universal LR model, termed Epitope Presentation Integrated Prediction (EPIP), based on more than 180,000 unique HLA ligands collected from public sources and ~3,000 HLA ligands generated by ourselves, to predict epitope presentation for 66 common HLA-I alleles. Results When evaluating EPIP on large, independent HLA eluted ligand datasets, it performed substantially better than other popular methods, including MixMHCpred (v2.0), NetMHCpan (v4.0), and MHCflurry (v1.2.2), with an average 0.1% positive predictive value (PPV) of 52.01%, compared to 37.24%, 36.96%, 24.90% and 23.76% achieved by MixMHCpred, NetMHCpan-4.0 (EL), NetMHCpan-4.0 (BA) and MHCflurry, respectively. It is also comparable to EDGE, a recent deep learning-based model that is not publicly available, on predicting epitope presentation and selecting immunogenic cancer neoantigens. However, the simplicity and flexibility of EPIP makes it easier to be applied in diverse situations, and we demonstrated this by generating MS data for the HCC4006 cell line and adding the support of HLA-A*33:03 to EPIP. EPIP is publicly available as a web tool < http://epip.genomics.cn/ >. Conclusions we have developed an easy to use, publicly available epitope prediction tool, EPIP, that incorporates information from both MS and RNA-seq data, and demonstrated its superior performance over existing public methods.