PocketVE improves 3D drug design stability with guided molecule generation
PocketVE: Stable and Property-Guided Structure-Based Drug Design with Variance-Exploding Diffusion
Machine Learning
Summary
Designing new drugs requires creating molecules that fit well inside a protein's pocket and have the right properties. The authors introduce PocketVE, a method that generates 3D molecules conditioned on protein pockets while controlling their physical stability and chemical features. PocketVE improves how well molecules fit and reduces unrealistic strain, making the designs more physically plausible. The method balances fitting the target pocket and preserving molecule shape by guiding the generation process during inference. This approach supports better drug-like molecule designs for future research and development.
What this means in practice
- •For drug discovery teams: Generate stable and property-optimized molecules tailored to protein binding pockets for improved candidate drugs.
- •For biopharmaceutical developers: Incorporate PocketVE’s framework to enhance molecule generation pipelines with better geometric stability and target compatibility.$Commercial implications: Enables commercial drug design platforms to offer improved structure-based molecule generation that reduces trial and error in lead discovery.
Authors
Peining Zhang, Jinbo Bi
Abstract
Protein-conditioned 3D molecule generation is a central challenge in structure-based drug design, requiring a balance between pocket compatibility, molecular properties, and physical geometry. We propose \textbf{PocketVE}, a protein-pocket-conditioned variance-exploding (VE) diffusion framework that couples stable coordinate denoising with inference-time property guidance. Specifically, PocketVE combines an EDM-style training and sampling setup for 3D denoising, classifier-free guidance for multi-property steering without external property classifiers, and adaptive protein perturbation as a training-time pocket regularizer. Evaluated on CrossDocked2020 under the GenBench3D protocol, PocketVE improves Valid$_{3\text{D}}$ from 58.6 to 80.6 and reduces strain energy from 457.4 to 127.9 relative to its TAGMol architectural baseline, while retaining competitive docking and molecular-property scores under moderate guidance. A guidance-scale study shows that moderate guidance gives a favorable balance between target-related objectives and geometric quality, whereas stronger guidance can degrade geometry and distributional fidelity. Pocket-permutation and PoseCheck diagnostics further support pocket-specific spatial compatibility with reduced steric conflicts. Overall, the results suggest that geometric stability and inference-time property guidance should be considered as coupled design objectives.