AbFlowNet: Optimizing Antibody-Antigen Binding Energy via Diffusion-GFlowNet Fusion

TL;DR

AbFlowNet introduces a novel generative framework combining GFlowNet and Diffusion models to optimize antibody binding energy efficiently, outperforming existing methods in amino acid recovery, geometric reconstruction, and energy metrics.

Contribution

The paper presents AbFlowNet, a new approach that unifies diffusion and reward optimization for antibody design, directly incorporating binding energy signals into the generative process.

Findings

Outperforms base diffusion model in amino acid recovery by 3.06%

Achieves 20.40% improvement in geometric reconstruction (RMSD)

Reduces binding energy errors by 38.1%

Abstract

Complementarity Determining Regions (CDRs) are critical segments of an antibody that facilitate binding to specific antigens. Current computational methods for CDR design utilize reconstruction losses and do not jointly optimize binding energy, a crucial metric for antibody efficacy. Rather, binding energy optimization is done through computationally expensive Online Reinforcement Learning (RL) pipelines rely heavily on unreliable binding energy estimators. In this paper, we propose AbFlowNet, a novel generative framework that integrates GFlowNet with Diffusion models. By framing each diffusion step as a state in the GFlowNet framework, AbFlowNet jointly optimizes standard diffusion losses and binding energy by directly incorporating energy signals into the training process, thereby unifying diffusion and reward optimization in a single procedure. Experimental results show that AbFlowNet outperforms the base diffusion model by 3.06% in amino acid recovery, 20.40% in geometric reconstruction (RMSD), and 3.60% in binding energy improvement ratio. ABFlowNet also decreases Top-1 total energy and binding energy errors by 24.8% and 38.1% without pseudo-labeling the test dataset or using computationally expensive online RL regimes.