Discovery of novel, high-affinity HBV T-cell epitopes by molecular dynamics-augmented proteome-wide screening

Accurate prediction of peptide-MHC (pMHC) binding is critical for vaccine design, yet conventional sequence-based methods alone lack the quantitative accuracy required to prioritize the most potent high-affinity epitopes. This creates a significant bottleneck in the development of T-cell-based immunotherapies. To address this, we established a sequential screening framework that first employs NetMHC-based prediction to rapidly identify potential HBV binders, followed by template-guided molecular dynamics (MD) refinement to evaluate their biophysical stability and eliminate false positives. This integrated approach enabled a proteome-wide discovery of high-confidence HBV T-cell epitopes with superior binding affinities. Subsequent experimental validation confirmed their exceptional potency, with a remarkable nine newly identified peptides exhibiting binding affinities superior to a clinically validated positive control. Notably, the HBsAg candidate FLGGTTVCL emerged as the top binder, demonstrating a relative affinity several-fold higher than the control, highlighting its significant therapeutic potential. The success of this discovery was enabled by the framework's ability to provide a multi-faceted biophysical assessment, accurately filtering false positives by identifying underlying dynamic instabilities or non-physiological binding modes, which dramatically increased screening precision. Our work not only delivers a set of high-value HBV vaccine candidates but also establishes a robust blueprint for the rational discovery of elite epitopes, accelerating the development of next-generation vaccines and immunotherapies.