Prior-Guided Flow Matching for Target-Aware Molecule Design with Learnable Atom Number

2025-09-01

AI Summary

In structure-based drug design, generated molecules often suffer from geometric mismatch with protein binding pockets, unstable binding poses, and inconsistent molecular weight. To address these challenges, this paper introduces PAFlow—a novel generative model that (1) pioneers conditional flow matching for discrete atom-type modeling, (2) integrates a protein–ligand interaction prediction network to guide conformational and chemical generation toward biologically plausible binding modes, and (3) incorporates a learnable pocket-aware atom-count predictor for adaptive molecular size control. PAFlow jointly enforces geometric, chemical, and biophysical constraints within a unified probabilistic framework. Evaluated on the CrossDocked2020 benchmark, it achieves state-of-the-art performance with a mean AutoDock Vina score of −8.31, demonstrating substantial improvements in predicted binding affinity and drug-likeness. The approach establishes a new paradigm for target-controllable, physics-informed generative drug design.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Structure-based drug design (SBDD), aiming to generate 3D molecules with high binding affinity toward target proteins, is a vital approach in novel drug discovery. Although recent generative models have shown great potential, they suffer from unstable probability dynamics and mismatch between generated molecule size and the protein pockets geometry, resulting in inconsistent quality and off-target effects. We propose PAFlow, a novel target-aware molecular generation model featuring prior interaction guidance and a learnable atom number predictor. PAFlow adopts the efficient flow matching framework to model the generation process and constructs a new form of conditional flow matching for discrete atom types. A protein-ligand interaction predictor is incorporated to guide the vector field toward higher-affinity regions during generation, while an atom number predictor based on protein pocket information is designed to better align generated molecule size with target geometry. Extensive experiments on the CrossDocked2020 benchmark show that PAFlow achieves a new state-of-the-art in binding affinity (up to -8.31 Avg. Vina Score), simultaneously maintains favorable molecular properties.

Problem

• Generates 3D molecules with high protein binding affinity
• Addresses unstable probability dynamics in molecular generation
• Solves molecule-protein geometry mismatch through learnable predictors

Innovation

• Flow matching framework for discrete atom generation
• Protein-ligand interaction predictor guides affinity
• Learnable atom number predictor aligns geometry