Machine learning approaches for interpretable antibody property prediction using structural data

2025-10-27

AI Summary

Understanding the sequence–structure–function relationships of antibodies and their underlying molecular mechanisms remains a key challenge in computational immunology. Method: We propose an interpretable graph neural network framework that encodes antibody conformations as residue-level graphs, jointly predicting global binding affinity and local residue flexibility. The model integrates two complementary tasks—ANTIPASTI for affinity modeling and INFUSSE for flexibility modeling—and incorporates explainability techniques, including gradient-weighted class activation mapping (Grad-CAM), to identify functionally critical residues and conformation-determining regions. Contribution/Results: Our method achieves state-of-the-art performance across multiple benchmark datasets. Crucially, it provides biologically grounded interpretability by revealing mechanistic insights into antibody–antigen interactions. This enables mechanism-driven rational antibody design, offering both theoretical foundations and practical tools for therapeutic antibody engineering.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Graph-based Machine Learning
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Understanding the relationship between antibody sequence, structure and function is essential for the design of antibody-based therapeutics and research tools. Recently, machine learning (ML) models mostly based on the application of large language models to sequence information have been developed to predict antibody properties. Yet there are open directions to incorporate structural information, not only to enhance prediction but also to offer insights into the underlying molecular mechanisms. This chapter provides an overview of these approaches and describes two ML frameworks that integrate structural data (via graph representations) with neural networks to predict properties of antibodies: ANTIPASTI predicts binding affinity (a global property) whereas INFUSSE predicts residue flexibility (a local property). We survey the principles underpinning these models; the ways in which they encode structural knowledge; and the strategies that can be used to extract biologically relevant statistical signals that can help discover and disentangle molecular determinants of the properties of interest.

Problem

• Predicting antibody properties using structural data and machine learning
• Integrating structural information to enhance antibody property prediction
• Developing interpretable models to understand antibody structure-function relationships

Innovation

• Integrates structural data via graph representations
• Uses neural networks to predict antibody properties
• Extracts biological signals from statistical models