GNINA 1.0: molecular docking with deep learning

By Andrew T. McNutt, Paul G. Francoeur, Rishal Aggarwal, Tomohide Masuda, Rocco Meli, Matthew Ragoza, Jocelyn Sunseri, David Ryan Koes. 2021

765 citations recorded by the source corpus.

Molecular docking computationally predicts the conformation of a small molecule when binding to a receptor. Scoring functions are a vital piece of any molecular docking pipeline as they determine the fitness of sampled poses. Here we describe and evaluate the 1.0 release of the Gnina docking software, which utilizes an ensemble of convolutional neural networks (CNNs) as a scoring function. We also explore an array of parameter values for Gnina 1.0 to optimize docking performance and computational cost. Docking performance, as evaluated by the percentage of targets where the top pose is better than 2Å root mean square deviation (Top1), is compared to AutoDock Vina scoring when utilizing explicitly defined binding pockets or whole protein docking. Gnina , utilizing a CNN scoring function to rescore the output poses, outperforms AutoDock Vina scoring on redocking and cross-docking tasks when the binding pocket is defined (Top1 increases from 58% to 73% and from 27% to 37%, respectively) and when the whole protein defines the binding pocket (Top1 increases from 31% to 38% and from 12% to 16%, respectively). The derived ensemble of CNNs generalizes to unseen proteins and ligands and produces scores that correlate well with the root mean square deviation to the known binding pose. We provide the 1.0 version of Gnina under an open source license for use as a molecular docking tool at https://github.com/gnina/gnina .

Subfields: Computational Theory and Mathematics, Molecular Biology, Organic Chemistry, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

In Organic Chemistry

Shigenori Tanaka

In Computational Chemistry & Drug Discovery

Citations in the corpus

End-to-end protein–ligand complex structure generation with diffusion-based generative models (2022, methodology): “Nevertheless, DPL outperformed both DPL+S and GNINA when the holo structures were unavailable (Figure 3B), as the performance of these structure-dependent methods was signiﬁcantly degraded using apo structures (Figure 3C).”
TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction (2022, methodology): “We also included four popular docking methods QVina-W, GINAMcNutt et al. [2021], SMINAKoes et al. [2013] and GLIDEFriesner et al. [2004] as listed in Stärk et al. [2022].”
DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking (2022, methodology): “We choose 2Å since much prior work considers poses with an RMSD less that 2Å as “good" or successful [25, 15, 16].”
EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction (2022, methodology): “…used in AutoDock Vina, enabling better ﬁne-tuning of predicted docked poses, as well as direct binding afﬁnity prediction from the 3D complex (McNutt et al., 2021; Fran-coeur et al., 2020; Ragoza et al., 2017; Wallach et al., 2015; Lim et al., 2019; Morrone et al., 2020; Jiang et al., 2021;…”
Independent SE(3)-Equivariant Models for End-to-End Rigid Protein Docking (2021, background): “These models are typically based on heavy candidate sampling and ranking (Trott and Olson, 2010; Koes et al., 2013; McNutt et al., 2021; Bao et al., 2021), being tailored for small drug-like ligands and often assuming known binding pocket.”
Equivariant 3D-conditional diffusion model for molecular linker design (2022, neutral): “For all the experiments discussed in the main text, we sampled with the same number of denoising steps T = 500 as used in training.”