12 September 2025 in bioRxiv
The ability of T cells to discriminate self from non-self peptides is a cornerstone of adaptive immunity, yet the structural principles guiding this process remain poorly defined. Here, we present a proof-of-concept study using physicochemical descriptors derived from images of peptide-MHC (pMHC) surfaces to explore patterns of immunogenicity. We modeled 353 pMHC complexes restricted to HLA-A*02:01, comprising 106 self-derived peptides from the HLA Ligand Atlas and 247 viral epitopes from the Immune Epitope Database. Electrostatic potentials of TCR-facing interfaces were computed and transformed into quantitative features across 92 immunologically relevant regions of interest. Unsupervised hierarchical clustering uncovered subsets of pMHCs forming self-only, viral-only, and mixed clusters, revealing conserved structural fingerprints linked to tolerance or immunogenicity. Heatmap analysis identified close viral-self proximities, highlighting candidate pairs potentially involved in molecular mimicry and autoimmune triggers. Building on these findings, we trained an XGBoost classifier that achieved strong performance (F1 score 0.87; AUC 0.84) in distinguishing self from viral peptides. To our knowledge, this is the first demonstration that structural and physicochemical fingerprints of pMHC complexes are sufficient to discriminate self from non-self. These results establish a framework for computational immunology and provide hypotheses for validation, with implications for autoimmunity, vaccine design, and immunotherapy.
22 July 2026 in SSRN Electronic Journal
RTRenata Fioravanti Tarabini
Renata Fioravanti Tarabini
APAlexandre dos Santos Pereira
Alexandre dos Santos Pereira
T-cell cross-reactivity occurs when distinct peptide-MHC complexes are recognized by the same T-cell population, a phenomenon with important implications for vaccine design, cancer immunotherapy, and off-target immune responses. Since experimental structural determination of peptide-MHC complexes remains costly and time-consuming, computational modeling offers a practical alternative for investigating candidate cross-reactive epitopes. However, pMHC modeling methods are usually evaluated by structural similarity to crystallographic references rather than by their ability to reproduce known immunological cross-reactivity patterns. In this study, we compared DockTope, APE-GEN 2.0, AlphaFold3, and PANDORA according to their capacity to reproduce experimentally described cross-reactivity webs among HLA-A*02:01-restricted nonameric epitopes. Modeled pMHC structures were visualized in GRASP 2.0, and RGB values representing electrostatic potential in TCR-contacting regions were extracted to generate pairwise electrostatic-distance matrices. Depending on dataset structure, method performance was evaluated using ROC-AUC, relative compactness scores, or normalized discounted cumulative gain. Across the majority of the article-derived epitope sets, DockTope showed the strongest overall agreement with expected cross-reactivity behavior, achieving the highest composite score and best mean rank. APE-GEN 2.0 showed intermediate and dataset-dependent performance, while AlphaFold3 performed strongly in positive-only compactness datasets but less consistently when discrimination from non-reactive comparators was required. PANDORA showed the weakest overall agreement under this workflow. These findings suggest that pMHC modeling tools differ substantially in their ability to preserve TCR-facing electrostatic relationships relevant to cross-reactive recognition. Benchmarking pMHC models against immunological behavior, rather than structural accuracy alone, may improve computational prioritization of cross-reactive epitopes.