EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

Hannes Stärk, Octavian Ganea, Lagnajit Pattanaik, Dr.Regina Barzilay, Tommi Jaakkola
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:20503-20521, 2022.

Abstract

Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this paradigm with EquiBind, an SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand’s bound pose and orientation. EquiBind achieves significant speed-ups and better quality compared to traditional and recent baselines. Further, we show extra improvements when coupling it with existing fine-tuning techniques at the cost of increased running time. Finally, we propose a novel and fast fine-tuning model that adjusts torsion angles of a ligand’s rotatable bonds based on closed form global minima of the von Mises angular distance to a given input atomic point cloud, avoiding previous expensive differential evolution strategies for energy minimization.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-stark22b, title = {{E}qui{B}ind: Geometric Deep Learning for Drug Binding Structure Prediction}, author = {St{\"a}rk, Hannes and Ganea, Octavian and Pattanaik, Lagnajit and Barzilay, Dr.Regina and Jaakkola, Tommi}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {20503--20521}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/stark22b/stark22b.pdf}, url = {https://proceedings.mlr.press/v162/stark22b.html}, abstract = {Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this paradigm with EquiBind, an SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand’s bound pose and orientation. EquiBind achieves significant speed-ups and better quality compared to traditional and recent baselines. Further, we show extra improvements when coupling it with existing fine-tuning techniques at the cost of increased running time. Finally, we propose a novel and fast fine-tuning model that adjusts torsion angles of a ligand’s rotatable bonds based on closed form global minima of the von Mises angular distance to a given input atomic point cloud, avoiding previous expensive differential evolution strategies for energy minimization.} }
Endnote
%0 Conference Paper %T EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction %A Hannes Stärk %A Octavian Ganea %A Lagnajit Pattanaik %A Dr.Regina Barzilay %A Tommi Jaakkola %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-stark22b %I PMLR %P 20503--20521 %U https://proceedings.mlr.press/v162/stark22b.html %V 162 %X Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this paradigm with EquiBind, an SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand’s bound pose and orientation. EquiBind achieves significant speed-ups and better quality compared to traditional and recent baselines. Further, we show extra improvements when coupling it with existing fine-tuning techniques at the cost of increased running time. Finally, we propose a novel and fast fine-tuning model that adjusts torsion angles of a ligand’s rotatable bonds based on closed form global minima of the von Mises angular distance to a given input atomic point cloud, avoiding previous expensive differential evolution strategies for energy minimization.
APA
Stärk, H., Ganea, O., Pattanaik, L., Barzilay, D. & Jaakkola, T.. (2022). EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:20503-20521 Available from https://proceedings.mlr.press/v162/stark22b.html.

Related Material