Deep generative design with 3D pharmacophoric constraints

Imrie, Fergus; Hadfield, Thomas E.; Bradley, A.R.; Deane, Charlotte M.

doi:10.1039/d1sc02436a

Cited by 60 publications

(68 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applications of drug binding methods. Computational docking methods are employed for various facets of drug discovery, e.g., fast virtual screening (Gniewek et al, 2021;Jastrzebski et al, 2020) or de novo binder generation (Masuda et al, 2020;Imrie et al, 2021;Drotár et al, 2021).…”

Section: Related Workmentioning

confidence: 99%

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

Stark¹,

Ganea²,

Pattanaik³

et al. 2022

Preprint

View full text Add to dashboard Cite

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.

show abstract

Section: Related Workmentioning

confidence: 99%

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

Stark¹,

Ganea²,

Pattanaik³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“… 39 Others used reinforcement learning to guide the sampling of 3D ligands towards high affinity for a target protein 30 or conditioned the generation of molecular graphs on density grids of 3D pharmacophores. 40 However, generating 3D molecular structures directly from protein binding pockets remains an unsolved challenge. 41 To address this, we make the following contributions:…”

Section: Introductionmentioning

confidence: 99%

Generating 3D molecules conditional on receptor binding sites with deep generative models

2022

View full text Add to dashboard Cite

show abstract

“…However, none of the above approaches allow for the specification of a preferred elaboration size, which, combined with their inability to account for protein structure when generating elaborations, means they cannot ensure that elaborations made by the model would be of an appropriate size to fit within the binding pocket. More recently, we proposed DEVELOP, 19 a fragment-based generative model for linking and growing which built on our DeLinker 20 model. DEVELOP allows the specification of pharmacophoric constraints and linker/elaboration length, providing a greater degree of control over the resulting molecules.…”

Section: Introductionmentioning

confidence: 99%

Incorporating Target-Specific Pharmacophoric Information into Deep Generative Models for Fragment Elaboration

Hadfield

Imrie

Merritt

et al. 2022

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Despite recent interest in deep generative models for scaffold elaboration, their applicability to fragment-to-lead campaigns has so far been limited. This is primarily due to their inability to account for local protein structure or a user’s design hypothesis. We propose a novel method for fragment elaboration, STRIFE, that overcomes these issues. STRIFE takes as input fragment hotspot maps (FHMs) extracted from a protein target and processes them to provide meaningful and interpretable structural information to its generative model, which in turn is able to rapidly generate elaborations with complementary pharmacophores to the protein. In a large-scale evaluation, STRIFE outperforms existing, structure-unaware, fragment elaboration methods in proposing highly ligand-efficient elaborations. In addition to automatically extracting pharmacophoric information from a protein target’s FHM, STRIFE optionally allows the user to specify their own design hypotheses.

show abstract

Deep generative design with 3D pharmacophoric constraints

Abstract: Generative models have increasingly been proposed as a solution to the molecular design problem. However, it has proved challenging to control the design process or incorporate prior knowledge, limiting their...

Cited by 60 publications

References 52 publications

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

Generating 3D molecules conditional on receptor binding sites with deep generative models

Incorporating Target-Specific Pharmacophoric Information into Deep Generative Models for Fragment Elaboration

Contact Info

Product

Resources

About