Structure-Regularized Interpretable TCR-Epitope Prediction
For immunology researchers, this work provides an interpretable model for TCR-epitope prediction and critically evaluates the limitations of current structure prediction methods in this domain.
TCR-SRIM, a structure-regularized interpretable model, achieves state-of-the-art predictive performance on the TCR-XAI benchmark for TCR-epitope binding prediction, and reveals that generated structures (e.g., from AlphaFold3) yield less accurate interaction patterns and reduced binding-site diversity compared to experimental structures.
T cell receptor (TCR)-epitope binding prediction is essential for understanding adaptive immunity and developing immunotherapies. Existing sequence- and structure-based models often generalize poorly to unseen epitopes and provide limited interpretability. Furthermore, the impact of generated structures on model learning remains unclear. We present TCR-SRIM, a structure-regularized interpretable-by-design model that combines protein language model embeddings with interpretable contact prototypes to capture residue-level TCR-epitope interactions. TCR-SRIM achieves state-of-the-art predictive performance and improved interpretation quality on the TCR-XAI benchmark. Using its inherent interpretability, we further evaluate the effect of generated structures on model learning. While structures predicted by AlphaFold3, TCRModel2, and tFold-TCR yield competitive performance, they lead to less accurate interaction patterns and reduced binding-site diversity than experimentally-resolved structures. Our results highlight limitations of current structure prediction models for TCR-epitope learning and demonstrate the value of interpretable-by-design models for studying generated biological structures.