FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction

Included are preprocessed datasets and model weights accompanying the manuscript "FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction" [1]. In particular, the preprocessed PDBBind-E dataset, which is comprised of apo (holo) predicted (crystal) protein structure pairs for PDBBind 2020, Binding MOAD, DockGen, and the PDB-based van der Mers (vdm) dataset, is available for download. Note that the included "holo_aligned" protein structures (predicted by ESMFold) for each constituent dataset have been pre-aligned w.r.t. the corresponding holo (crystal) protein structures.

Paper Abstract:

Motivation Powerful generative AI models of protein-ligand structure have recently been proposed, but few of these methods support both flexible protein-ligand docking and affinity estimation. Of those that do, none can directly model multiple binding ligands concurrently or have been rigorously benchmarked on pharmacologically relevant drug targets, hindering their widespread adoption in drug discovery efforts.
Results In this work, we propose FlowDock, the first deep geometric generative model based on conditional flow matching that learns to directly map unbound (apo) structures to their bound (holo) counterparts for an arbitrary number of binding ligands. Furthermore, FlowDock provides predicted structural confidence scores and binding affinity values with each of its generated protein-ligand complex structures, enabling fast virtual screening of new (multi-ligand) drug targets. For the well-known PoseBusters Benchmark dataset, FlowDock outperforms single-sequence AlphaFold 3 with a 51% blind docking success rate using unbound (apo) protein input structures and without any information derived from multiple sequence alignments, and for the challenging new DockGen-E dataset, FlowDock outperforms single-sequence AlphaFold 3 and matches single-sequence Chai-1 for binding pocket generalization. Additionally, in the ligand category of the 16th community-wide Critical Assessment of Techniques for Structure Prediction (CASP16), FlowDock ranked among the top-5 methods for pharmacological binding affinity estimation across 140 protein-ligand complexes, demonstrating the efficacy of its learned representations in virtual screening.
Availability and implementation Source code, data, and pre-trained models are available at https://github.com/BioinfoMachineLearning/FlowDock.

References

[1] Morehead A, Cheng J. FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction. arXiv; 2024. Available from: https://arxiv.org/abs/2412.10966 (ISMB 2025)