We've been investigating deep learning-based protein-ligand docking methods which often claim to be able to generate ligand binding modes within 2Å RMSD of the experimental one. We found, however, this simple criterion can conceal a multitude of chemical and structural sins.
1/4
If you're interested in assessing the structural quality and chemical validity of predicted binding modes (and conformations) of small molecules, you might like to read about one of our DPhil students Martin Buttenschoen's work on PoseBusters:
https://arxiv.org/abs/2308.05777
2/4
The last few years have seen the development of numerous deep learning-based protein-ligand docking methods. They offer huge promise in terms of speed and accuracy. However, despite claims of state-of-the-art performance in terms of crystallographic root-mean-square deviation (RMSD), upon closer inspection, it has become apparent that they often produce physically implausible molecular structures. It is therefore not sufficient to evaluate these methods solely by RMSD to a native binding mode. It is vital, particularly for deep learning-based methods, that they are also evaluated on steric and energetic criteria. We present PoseBusters, a Python package that performs a series of standard quality checks using the well-established cheminformatics toolkit RDKit. Only methods that both pass these checks and predict native-like binding modes should be classed as having "state-of-the-art" performance. We use PoseBusters to compare five deep learning-based docking methods (DeepDock, DiffDock, EquiBind, TankBind, and Uni-Mol) and two well-established standard docking methods (AutoDock Vina and CCDC Gold) with and without an additional post-prediction energy minimisation step using a molecular mechanics force field. We show that both in terms of physical plausibility and the ability to generalise to examples that are distinct from the training data, no deep learning-based method yet outperforms classical docking tools. In addition, we find that molecular mechanics force fields contain docking-relevant physics missing from deep-learning methods. PoseBusters allows practitioners to assess docking and molecular generation methods and may inspire new inductive biases still required to improve deep learning-based methods, which will help drive the development of more accurate and more realistic predictions.
Martin has developed a pip-installable Python package that's easy to use:
https://github.com/maabuu/posebusters
3/4
You can also hear Martin speak at the upcoming RSC CICAG and RSC BMRC's 6th "AI in Chemistry" Symposium at Churchill College, Cambridge—and there's still time to register, too!
https://rscbmcs.org/events/aichem23/
4/4
gmm
