Guided Sequence-Structure Generative Modeling for Iterative Antibody Optimization

2025-09-19

AI Summary

Antibody preclinical optimization typically requires multiple iterative engineering cycles, yet existing methods struggle to consistently incorporate structural priors across iterations due to the scarcity of experimental complex structures. To address this, we present the first structure-aware iterative optimization framework for antibody–antigen complexes. Our approach integrates a sequence-structure joint diffusion generative model with end-to-end complex structure prediction and experimental feedback-guided sampling—enabling continuous structure-aware design without requiring full experimental structures at each round. Crucially, it supports multi-round co-optimization of affinity and developability without periodic structural re-determination. In silico simulations and in vitro experiments across multiple targets demonstrate rapid generation of high-affinity antibodies, reducing optimization cycles by over 40% and significantly improving key developability metrics—including solubility, thermal stability, and aggregation resistance.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Optimization
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Therapeutic antibody candidates often require extensive engineering to improve key functional and developability properties before clinical development. This can be achieved through iterative design, where starting molecules are optimized over several rounds of in vitro experiments. While protein structure can provide a strong inductive bias, it is rarely used in iterative design due to the lack of structural data for continually evolving lead molecules over the course of optimization. In this work, we propose a strategy for iterative antibody optimization that leverages both sequence and structure as well as accumulating lab measurements of binding and developability. Building on prior work, we first train a sequence-structure diffusion generative model that operates on antibody-antigen complexes. We then outline an approach to use this model, together with carefully predicted antibody-antigen complexes, to optimize lead candidates throughout the iterative design process. Further, we describe a guided sampling approach that biases generation toward desirable properties by integrating models trained on experimental data from iterative design. We evaluate our approach in multiple in silico and in vitro experiments, demonstrating that it produces high-affinity binders at multiple stages of an active antibody optimization campaign.

Problem

• Optimizing therapeutic antibodies lacking structural data during iterative design
• Leveraging sequence-structure modeling for antibody-antigen complex optimization
• Improving antibody binding affinity through guided generative modeling

Innovation

• Combines sequence-structure diffusion generative modeling
• Uses predicted antibody-antigen complexes for optimization
• Guides sampling with models trained on experimental data