English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The last few years have seen the development of numerous deep learning-basedprotein-ligand docking methods. They offer huge promise in terms of speed andaccuracy. However, despite claims of state-of-the-art performance in terms ofcrystallographic root-mean-square deviation (RMSD), upon closer inspection, ithas become apparent that they often produce physically implausible molecularstructures. It is therefore not sufficient to evaluate these methods solely byRMSD to a native binding mode. It is vital, particularly for deeplearning-based methods, that they are also evaluated on steric and energeticcriteria. We present PoseBusters, a Python package that performs a series ofstandard quality checks using the well-established cheminformatics toolkitRDKit. Only methods that both pass these checks and predict native-like bindingmodes should be classed as having "state-of-the-art" performance. We usePoseBusters to compare five deep learning-based docking methods (DeepDock,DiffDock, EquiBind, TankBind, and Uni-Mol) and two well-established standarddocking methods (AutoDock Vina and CCDC Gold) with and without an additionalpost-prediction energy minimisation step using a molecular mechanics forcefield. We show that both in terms of physical plausibility and the ability togeneralise to examples that are distinct from the training data, no deeplearning-based method yet outperforms classical docking tools. In addition, wefind that molecular mechanics force fields contain docking-relevant physicsmissing from deep-learning methods. PoseBusters allows practitioners to assessdocking and molecular generation methods and may inspire new inductive biasesstill required to improve deep learning-based methods, which will help drivethe development of more accurate and more realistic predictions.

PoseBusters: AI-based docking methods fail to generate physically valid poses or generalise to novel sequences