Piloting Structure-Based Drug Design via Modality-Specific Optimal Schedule

TL;DR

This paper introduces a novel VLB-Optimal Scheduling strategy for structure-based drug design that improves molecular geometry modeling and achieves state-of-the-art performance in pose prediction accuracy.

Contribution

It proposes a new VLB-Optimal Scheduling method to optimize the probability path in multi-modal molecular structure modeling, enhancing drug design accuracy.

Findings

Achieved 95.9% PoseBusters passing rate on CrossDock

Improved baseline performance by over 10%

Maintained high affinity and validity in molecular predictions

Abstract

Structure-Based Drug Design (SBDD) is crucial for identifying bioactive molecules. Recent deep generative models are faced with challenges in geometric structure modeling. A major bottleneck lies in the twisted probability path of multi-modalities -- continuous 3D positions and discrete 2D topologies -- which jointly determine molecular geometries. By establishing the fact that noise schedules decide the Variational Lower Bound (VLB) for the twisted probability path, we propose VLB-Optimal Scheduling (VOS) strategy in this under-explored area, which optimizes VLB as a path integral for SBDD. Our model effectively enhances molecular geometries and interaction modeling, achieving state-of-the-art PoseBusters passing rate of 95.9% on CrossDock, more than 10% improvement upon strong baselines, while maintaining high affinities and robust intramolecular validity evaluated on held-out test set. Code is available at https://github.com/AlgoMole/MolCRAFT.