BConformeR: A Conformer Based on Mutual Sampling for Unified Prediction of Continuous and Discontinuous Antibody Binding Sites

2025-08-16

AI Summary

Accurate prediction of both linear and conformational epitopes on antigens is critical for vaccine design, antibody development, and immunological mechanism studies; however, existing methods exhibit limited performance—particularly for conformational epitope identification. To address this, we propose Conformer, the first unified deep learning framework for jointly modeling both epitope types. It integrates convolutional neural networks (CNNs) to capture local sequence patterns and a Transformer architecture to encode long-range residue dependencies, augmented by a novel mutual-sampling training strategy to enhance optimization stability. Evaluated on a benchmark dataset of 1,080 antigen–antibody complexes, Conformer achieves statistically significant improvements over state-of-the-art methods across all major metrics: Pearson correlation coefficient (PCC), ROC-AUC, PR-AUC, and F1-score—with particularly notable gains in conformational epitope prediction accuracy. This work establishes a new paradigm for epitope prediction, offering both theoretical advances in multimodal representation learning and practical utility in immunoinformatics.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate prediction of antibody-binding sites (epitopes) on antigens is crucial for vaccine design, immunodiagnostics, therapeutic antibody development, antibody engineering, research into autoimmune and allergic diseases, and for advancing our understanding of immune responses. Despite in silico methods that have been proposed to predict both linear (continuous) and conformational (discontinuous) epitopes, they consistently underperform in predicting conformational epitopes. In this work, we propose a conformer-based model trained on antigen sequences derived from 1,080 antigen-antibody complexes, leveraging convolutional neural networks (CNNs) to extract local features and Transformers to capture long-range dependencies within antigen sequences. Ablation studies demonstrate that CNN enhances the prediction of linear epitopes, and the Transformer module improves the prediction of conformational epitopes. Experimental results show that our model outperforms existing baselines in terms of PCC, ROC-AUC, PR-AUC, and F1 scores on conformational epitopes.

Problem

• Predicting antibody-binding sites on antigens accurately
• Improving prediction of discontinuous conformational epitopes
• Unifying prediction for continuous and discontinuous epitopes

Innovation

• Conformer model for epitope prediction
• Combines CNN and Transformer features
• Outperforms in conformational epitope prediction