DecoderTCR: Compositional Pretraining and Entropy-Guided Decoding for TCR-pMHC Interactions
Authors:
Preprint
Abstract
Modeling recognition between T-cell receptors (TCRs) and peptide-MHC (pMHC) complexes is a fundamental challenge in computational immunology, constrained by sparse paired interaction data relative to abundant unpaired sequences. We introduce DecoderTCR, a masked language model framework that addresses this through two contributions: (1) a compositional continual pre-training curriculum that learns component representations from marginal data before refining cross-chain dependencies from limited pairs, and (2) Iterative Entropy-Guided Refinement (IEGR), a non-autoregressive decoding algorithm that resolves high-confidence positions first to provide context for uncertain regions. On held-out benchmarks, DecoderTCR achieves 0.96 AUROC for zero-shot pMHC binding prediction and 0.76 AUROC for epitope-specific TCR recognition, approaching supervised baselines without epitope-specific training. Learned representations recover structural contacts without coordinate supervision, and generated sequences exhibit realistic recombination statistics. Experimental validation reveals a prediction-generation gap: strong discrimination does not yet yield reliable generation, highlighting an open challenge for the field.Keywords
Funding Information
- NIH (DP2AI177884)
This publication has 43 references indexed in Scilit:
- The Observed T Cell Receptor Space database enables paired-chain repertoire mining, coherence analysis, and language modelingCell Reports, 2024
- IgLM: Infilling language modeling for antibody sequence designCell Systems, 2023
- Improved predictions of antigen presentation and TCR recognition with MixMHCpred2.2 and PRIME2.0 reveal potent SARS-CoV-2 CD8+ T-cell epitopesCell Systems, 2023
- ProtGPT2 is a deep unsupervised language model for protein designNature Communications, 2022
- Precision Neoantigen Discovery Using Large-scale Immunopeptidomes and Composite Modeling of MHC Peptide PresentationMolecular & Cellular Proteomics, 2021
- MHCflurry 2.0: Improved Pan-Allele Prediction of MHC Class I-Presented Peptides by Incorporating Antigen ProcessingCell Systems, 2020
- Quantifiable predictive features define epitope-specific T cell receptor repertoiresNature, 2017
- Mass Spectrometry Profiling of HLA-Associated Peptidomes in Mono-allelic Cells Enables More Accurate Epitope PredictionImmunity, 2017
- MHC class II proteins and disease: a structural perspectiveNature Reviews Immunology, 2006
- The antigenic identity of peptide-MHC complexes: A comparison of the conformations of five viral peptides presented by HLA-A2Cell, 1993