DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

By Gabriele Corso, Hannes Stärk, Bowen Jing, Regina Barzilay, Tommi S. Jaakkola. 2022

790 citations recorded by the source corpus.

Predicting the binding structure of a small molecule ligand to a protein -- a task known as molecular docking -- is critical to drug design. Recent deep learning methods that treat docking as a regression problem have decreased runtime compared to traditional search-based methods but have yet to offer substantial improvements in accuracy. We instead frame molecular docking as a generative modeling problem and develop DiffDock, a diffusion generative model over the non-Euclidean manifold of ligand poses. To do so, we map this manifold to the product space of the degrees of freedom (translational, rotational, and torsional) involved in docking and develop an efficient diffusion process on this space. Empirically, DiffDock obtains a 38% top-1 success rate (RMSD<2A) on PDBBind, significantly outperforming the previous state-of-the-art of traditional docking (23%) and deep learning (20%) methods. Moreover, while previous methods are not able to dock on computationally folded structures (maximum accuracy 10.4%), DiffDock maintains significantly higher precision (21.7%). Finally, DiffDock has fast inference times and provides confidence estimates with high selective accuracy.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology, Generative Models, 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 Materials Chemistry

In Molecular Biology

In Generative Models

In Computational Chemistry & Drug Discovery

Citations in the corpus

End-to-end protein–ligand complex structure generation with diffusion-based generative models (2022, methodology): “To address this issue, we could train a new scoring model as in DiffDock [35] or exploit likelihood estimation techniques using diffusion-based generative models [15,29].”
Boltz-1 Democratizing Biomolecular Interaction Modeling (2024, positive): “In particular, the use of deep generative models has proven to be effective in modeling the interaction between different biomolecules with DiffDock [Corso et al., 2022] showing significant improvements over traditional molecular docking approaches and, most recently, AlphaFold3 [Abramson et al.,…”
Do Deep Learning Models Really Outperform Traditional Approaches in Molecular Docking? (2023, methodology): “We directly use the number reported in DiffDock [12] paper.”
SE(3) diffusion model with application to protein backbone generation (2023, methodology): “consider the product of tori for molecular conformer generation, Corso et al. (2023) on the product space R(3) × SO(3) × SO(2) for protein docking applications and Leach et al.”
Equivariant 3D-conditional diffusion model for molecular linker design (2022, neutral): “More details on datasets, baselines and metrics can be found in Methods.”
Improving and generalizing flow-based generative models with minibatch optimal transport (2023, positive): “…diffusion models, which are the current state of the art on many generative modeling tasks (Dhariwal & Nichol, 2021; Austin et al., 2021; Corso et al., 2022; Watson et al., 2022b), approximate a stochastic differential equation (SDE) that transforms a simple density to the data…”
Structure-based drug design with equivariant diffusion models (2022, positive): “Despite remarkable progress in the use of deep learning as surrogate docking models (Lu et al., 2022; Stärk et al., 2022; Corso et al., 2022), deep learning-based design of ligands that bind to target proteins remains an open problem.”
Generating Molecular Conformer Fields (2023, methodology): “These approaches tend to use equivariant network architectures to deal with molecular graphs (Xu et al., 2022) or model domain-specific factors like torsional angles (Ganea et al., 2021; Jing et al., 2022).”