RapidDock: Unlocking Proteome-scale Molecular Docking

TL;DR

RapidDock is a transformer-based model that significantly accelerates molecular docking, enabling large-scale drug discovery with high accuracy and minimal inference time.

Contribution

We introduce RapidDock, a novel transformer-based approach that achieves over 100x speedup in molecular docking without sacrificing accuracy.

Findings

Achieves 52.1% and 44.0% success rates on benchmarks.

Inference time of 0.04 seconds per prediction on a single GPU.

Provides key architectural insights for transformer use in molecular docking.

Abstract

Accelerating molecular docking -- the process of predicting how molecules bind to protein targets -- could boost small-molecule drug discovery and revolutionize medicine. Unfortunately, current molecular docking tools are too slow to screen potential drugs against all relevant proteins, which often results in missed drug candidates or unexpected side effects occurring in clinical trials. To address this gap, we introduce RapidDock, an efficient transformer-based model for blind molecular docking. RapidDock achieves at least a $100 \times$ speed advantage over existing methods without compromising accuracy. On the Posebusters and DockGen benchmarks, our method achieves $52.1\%$ and $44.0\%$ success rates ($\text{RMSD}<2$\r{A}), respectively. The average inference time is $0.04$ seconds on a single GPU, highlighting RapidDock's potential for large-scale docking studies. We examine the key features of RapidDock that enable leveraging the transformer architecture for molecular docking, including the use of relative distance embeddings of $3$D structures in attention matrices, pre-training on protein folding, and a custom loss function invariant to molecular symmetries.