Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DapPep: Domain Adaptive Peptide-agnostic Learning for Universal T-cell Receptor-antigen Binding Affinity Prediction

Jiangbin Zheng, Qianhui Xu, Ruichen Xia, Stan Z. Li

TL;DR

This work introduces a domain-adaptive peptide-agnostic learning framework DapPep for universal TCR-antigen binding affinity prediction, which proves effective in challenging clinical tasks such as sorting reactive T cells in tumor neoantigen therapy and identifying key positions in 3D structures.

Abstract

Identifying T-cell receptors (TCRs) that interact with antigenic peptides provides the technical basis for developing vaccines and immunotherapies. The emergent deep learning methods excel at learning antigen binding patterns from known TCRs but struggle with novel or sparsely represented antigens. However, binding specificity for unseen antigens or exogenous peptides is critical. We introduce a domain-adaptive peptide-agnostic learning framework DapPep for universal TCR-antigen binding affinity prediction to address this challenge. The lightweight self-attention architecture combines a pre-trained protein language model with an inner-loop self-supervised regime to enable robust TCR-peptide representations. Extensive experiments on various benchmarks demonstrate that DapPep consistently outperforms existing tools, showcasing robust generalization capability, especially for data-scarce settings and unseen peptides. Moreover, DapPep proves effective in challenging clinical tasks such as sorting reactive T cells in tumor neoantigen therapy and identifying key positions in 3D structures.

DapPep: Domain Adaptive Peptide-agnostic Learning for Universal T-cell Receptor-antigen Binding Affinity Prediction

TL;DR

This work introduces a domain-adaptive peptide-agnostic learning framework DapPep for universal TCR-antigen binding affinity prediction, which proves effective in challenging clinical tasks such as sorting reactive T cells in tumor neoantigen therapy and identifying key positions in 3D structures.

Abstract

Identifying T-cell receptors (TCRs) that interact with antigenic peptides provides the technical basis for developing vaccines and immunotherapies. The emergent deep learning methods excel at learning antigen binding patterns from known TCRs but struggle with novel or sparsely represented antigens. However, binding specificity for unseen antigens or exogenous peptides is critical. We introduce a domain-adaptive peptide-agnostic learning framework DapPep for universal TCR-antigen binding affinity prediction to address this challenge. The lightweight self-attention architecture combines a pre-trained protein language model with an inner-loop self-supervised regime to enable robust TCR-peptide representations. Extensive experiments on various benchmarks demonstrate that DapPep consistently outperforms existing tools, showcasing robust generalization capability, especially for data-scarce settings and unseen peptides. Moreover, DapPep proves effective in challenging clinical tasks such as sorting reactive T cells in tumor neoantigen therapy and identifying key positions in 3D structures.

Paper Structure

This paper contains 10 sections, 4 equations, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Inner-loop TCR-Peptide Module Pre-training
  4. Binding Affinity Prediction Pipeline
  5. Experiments
  6. Setups
  7. DapPep Outperforms Baselines Across Different Settings
  8. DapPep Specializes in Unseen Peptides
  9. Validation for Potential Clinical Applications
  10. Conclusions and Limitations

Figures (2)

  • Figure 1: Proposed DapPep for TCR-peptide binding affinity prediction. The training pipeline is divided into two stages: Stage 1 involves initializing the TCR representation module (A) and pre-training the peptide prior module (B, C), while Stage 2 entails the pre-trained modules transferred from Stage 1 to optimize the overall framework (D). (E). Linear binding affinity decoder. (F). TCR-peptide cross-attention module. (G). Inference regimes for different data settings.
  • Figure 2: Comparison to SOTA model (PanPep) of ROC-AUC and PR-AUC performances in the different settings and datasets.