Learning from B Cell Evolution: Adaptive Multi-Expert Diffusion for Antibody Design via Online Optimization

2025-07-25

AI Summary

Existing antibody design methods lack adaptive antigen specificity. To address this, we propose the first online meta-learning antibody generation framework based on SE(3)-equivariant diffusion, inspired by B-cell affinity maturation. Our method integrates physical priors—including van der Waals interactions, molecular recognition, energy balance, and interface geometry—into a multi-expert co-evolutionary system. It further incorporates iterative feedback and symmetry-preserving mechanisms to enable dynamic, context-aware adaptation of generation strategies. The framework supports multi-objective trade-offs and patient-specific optimization. Empirical evaluation demonstrates significant improvements in hotspot residue coverage and binding interface quality across diverse targets—from small epitopes to large protein complexes—while exhibiting strong generalization. The resulting antibodies meet therapeutic-grade design criteria.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Online Learning & Bandits
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Recent advances in diffusion models have shown remarkable potential for antibody design, yet existing approaches apply uniform generation strategies that cannot adapt to each antigen's unique requirements. Inspired by B cell affinity maturation, where antibodies evolve through multi-objective optimization balancing affinity, stability, and self-avoidance, we propose the first biologically-motivated framework that leverages physics-based domain knowledge within an online meta-learning system. Our method employs multiple specialized experts (van der Waals, molecular recognition, energy balance, and interface geometry) whose parameters evolve during generation based on iterative feedback, mimicking natural antibody refinement cycles. Instead of fixed protocols, this adaptive guidance discovers personalized optimization strategies for each target. Our experiments demonstrate that this approach: (1) discovers optimal SE(3)-equivariant guidance strategies for different antigen classes without pre-training, preserving molecular symmetries throughout optimization; (2) significantly enhances hotspot coverage and interface quality through target-specific adaptation, achieving balanced multi-objective optimization characteristic of therapeutic antibodies; (3) establishes a paradigm for iterative refinement where each antibody-antigen system learns its unique optimization profile through online evaluation; (4) generalizes effectively across diverse design challenges, from small epitopes to large protein interfaces, enabling precision-focused campaigns for individual targets.

Problem

• Adaptive antibody design for unique antigen requirements
• Multi-objective optimization balancing affinity and stability
• Online meta-learning with physics-based domain knowledge

Innovation

• Adaptive multi-expert diffusion with online optimization
• Specialized experts evolve via iterative feedback cycles
• Target-specific adaptation without pre-training preserves symmetries