EMBER: Predicting TCR--pMHC Binding Without a Docked Complex
Abstract
We propose EMBER, a structure-based framework for TCR--pMHC binding prediction that eliminates the need for a predicted docked complex, and with it the placement error that dominates such predictions. Traditional structure-based predictors utilize the co-folded complex structure of the receptor and the antigen and interpret binding from its contact surface. Consequently, every downstream feature inherits any deviation the structure-prediction tool introduced. Methodologically, EMBER folds the TCR and pMHC independently, encodes them as SE(3)-invariant residue graphs, and compares their learned representations without constructing a shared coordinate frame. A relative pose is therefore not an input of the model, so placement error cannot propagate into the prediction. Computationally, EMBER preprocesses O(N+M) monomer folds where a complex-based predictor needs O(NM) operations. Experimentally, under epitope-disjoint evaluation with six retrained baselines, EMBER is first on every metric under all four folding tools, by 0.9 to 14.4% in micro-AUROC, 18.5 to 25.3% in micro-AUPRC and 10.6 to 24.1% in MCC. Its ranking is also up to 85% more stable across folding tools.
Related
- No Strings Attached: Predicting Tricuspid Valve Deformation Without In Vivo Chordal Geometry
- MPFIL: Multi-Level Protein Feature Integration for Predicting Mutation-Induced Protein-Protein Binding Affinity Changes
- CaliPPer: quantifying, predicting and improving AI model performance for binding prediction
- The ssDNA-binding protein MEIOB sustains synaptonemal complex integrity by stabilizing recombination intermediates during male meiosis
- Predicting single mutation effects on binding affinity at protein-protein interfaces based on MMPBSA calculations
BibTeX
@misc{ha2026ember,
title = {{EMBER: Predicting TCR--pMHC Binding Without a Docked Complex}},
author = {Lin Ha and Chen Zhou and Haozheng Luo and Han Liu and Xiaoyu Zhang and Yan Chen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.27334},
eprint = {2610.27334},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.27334},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - EMBER: Predicting TCR--pMHC Binding Without a Docked Complex
AU - Lin Ha
AU - Chen Zhou
AU - Haozheng Luo
AU - Han Liu
AU - Xiaoyu Zhang
AU - Yan Chen
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.27334
UR - https://arxiv.org/abs/2610.27334
AB - We propose EMBER, a structure-based framework for TCR--pMHC binding prediction that eliminates the need for a predicted docked complex, and with it the placement error that dominates such predictions. Traditional structure-based predictors utilize the co-folded complex structure of the receptor and the antigen and interpret binding from its contact surface. Consequently, every downstream feature inherits any deviation the structure-prediction tool introduced. Methodologically, EMBER folds the TCR and pMHC independently, encodes them as SE(3)-invariant residue graphs, and compares their learned representations without constructing a shared coordinate frame. A relative pose is therefore not an input of the model, so placement error cannot propagate into the prediction. Computationally, EMBER preprocesses O(N+M) monomer folds where a complex-based predictor needs O(NM) operations. Experimentally, under epitope-disjoint evaluation with six retrained baselines, EMBER is first on every metric under all four folding tools, by 0.9 to 14.4% in micro-AUROC, 18.5 to 25.3% in micro-AUPRC and 10.6 to 24.1% in MCC. Its ranking is also up to 85% more stable across folding tools.
ER -