Re-Dock: Towards Flexible and Realistic Molecular Docking with Diffusion Bridge

Yufei Huang, Odin Zhang, Lirong Wu, Cheng Tan, Haitao Lin, Zhangyang Gao, Siyuan Li, Stan Z. Li
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:20474-20489, 2024.

Abstract

Accurate prediction of protein-ligand binding structures, a task known as molecular docking is crucial for drug design but remains challenging. While deep learning has shown promise, existing methods often depend on holo-protein structures (docked, and not accessible in realistic tasks) or neglect pocket sidechain conformations, leading to limited practical utility and unrealistic conformation predictions. To fill these gaps, we introduce an under-explored task, named flexible docking to predict poses of ligand and pocket sidechains simultaneously and introduce Re-Dock, a novel diffusion bridge generative model extended to geometric manifolds. Specifically, we propose energy-to-geometry mapping inspired by the Newton-Euler equation to co-model the binding energy and conformations for reflecting the energy-constrained docking generative process. Comprehensive experiments on designed benchmark datasets including apo-dock and cross-dock demonstrate our model’s superior effectiveness and efficiency over current methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-huang24ag, title = {Re-Dock: Towards Flexible and Realistic Molecular Docking with Diffusion Bridge}, author = {Huang, Yufei and Zhang, Odin and Wu, Lirong and Tan, Cheng and Lin, Haitao and Gao, Zhangyang and Li, Siyuan and Li, Stan Z.}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {20474--20489}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/huang24ag/huang24ag.pdf}, url = {https://proceedings.mlr.press/v235/huang24ag.html}, abstract = {Accurate prediction of protein-ligand binding structures, a task known as molecular docking is crucial for drug design but remains challenging. While deep learning has shown promise, existing methods often depend on holo-protein structures (docked, and not accessible in realistic tasks) or neglect pocket sidechain conformations, leading to limited practical utility and unrealistic conformation predictions. To fill these gaps, we introduce an under-explored task, named flexible docking to predict poses of ligand and pocket sidechains simultaneously and introduce Re-Dock, a novel diffusion bridge generative model extended to geometric manifolds. Specifically, we propose energy-to-geometry mapping inspired by the Newton-Euler equation to co-model the binding energy and conformations for reflecting the energy-constrained docking generative process. Comprehensive experiments on designed benchmark datasets including apo-dock and cross-dock demonstrate our model’s superior effectiveness and efficiency over current methods.} }
Endnote
%0 Conference Paper %T Re-Dock: Towards Flexible and Realistic Molecular Docking with Diffusion Bridge %A Yufei Huang %A Odin Zhang %A Lirong Wu %A Cheng Tan %A Haitao Lin %A Zhangyang Gao %A Siyuan Li %A Stan Z. Li %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-huang24ag %I PMLR %P 20474--20489 %U https://proceedings.mlr.press/v235/huang24ag.html %V 235 %X Accurate prediction of protein-ligand binding structures, a task known as molecular docking is crucial for drug design but remains challenging. While deep learning has shown promise, existing methods often depend on holo-protein structures (docked, and not accessible in realistic tasks) or neglect pocket sidechain conformations, leading to limited practical utility and unrealistic conformation predictions. To fill these gaps, we introduce an under-explored task, named flexible docking to predict poses of ligand and pocket sidechains simultaneously and introduce Re-Dock, a novel diffusion bridge generative model extended to geometric manifolds. Specifically, we propose energy-to-geometry mapping inspired by the Newton-Euler equation to co-model the binding energy and conformations for reflecting the energy-constrained docking generative process. Comprehensive experiments on designed benchmark datasets including apo-dock and cross-dock demonstrate our model’s superior effectiveness and efficiency over current methods.
APA
Huang, Y., Zhang, O., Wu, L., Tan, C., Lin, H., Gao, Z., Li, S. & Li, S.Z.. (2024). Re-Dock: Towards Flexible and Realistic Molecular Docking with Diffusion Bridge. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:20474-20489 Available from https://proceedings.mlr.press/v235/huang24ag.html.

Related Material