Related Experiment Video
Updated: Jun 7, 2026
19:05
Measuring TCR-pMHC Binding In Situ using a FRET-based Microscopy Assay
Published on: October 30, 2015
Benchmarking TCR-pMHC structure prediction: a unified evaluation and CDR3-based functional insights
Jiadong Lu1,2, Xinyuan Zhu3, Xinting Hu1
1School of Artificial Intelligence and Data Science, University of Science and Technology of China, Huangshan Road, Shushan District, Hefei 230027, Anhui, China.
Briefings in Bioinformatics
|June 5, 2026
Summary
This study benchmarks T cell receptor (TCR) and peptide-major histocompatibility complex (pMHC) structure prediction models. MSA-based methods, particularly AlphaFold3, show superior accuracy, with the TCR CDR3 pLDDT score predicting structural and functional relevance.
Area of Science:
- Immunology
- Structural Biology
- Computational Biology
Background:
- T cell receptor (TCR) and peptide-major histocompatibility complex (pMHC) interactions are crucial for adaptive immunity.
- Accurate atomic-level modeling of these interactions is vital but lacks systematic evaluation of prediction tools.
Purpose of the Study:
- To benchmark various T cell receptor-peptide-major histocompatibility complex (TCR-pMHC) structure prediction models.
- To identify the most accurate and reliable methods for modeling these critical immune interactions.
Main Methods:
- A comprehensive benchmark was conducted using 70 previously unseen TCR-pMHC complexes.
- 13 structure prediction models were evaluated, including Multiple Sequence Alignment (MSA)-based, Predicted Local Distance Difference (PLD)-based, and docking-based approaches.
Main Results:
- MSA-based methods, especially AlphaFold3, demonstrated superior modeling accuracy and docking quality.
- The p-value of the difference in the TCR CDR3 region was identified as a reliable indicator of structural correctness and functional relevance.
- Reranking using the TCR CDR3 pLDDT score improved Top-1 success by up to 4.3% and captured 75.3% of mutation-induced affinity changes.
Conclusions:
- MSA-based approaches, particularly AlphaFold3, are the most effective for TCR-pMHC structure prediction.
- The TCR CDR3 pLDDT score enhances the practical utility of predicted structures for assessing functional relevance and guiding further research.
