Docking-based generative approaches in the search for new drug candidates

Danel, Tomasz; Łęski, Jan; Podlewska, Sabina; Podolak, Igor T.

doi:10.1016/j.drudis.2022.103439

Cited by 27 publications

(20 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These models are essentially ligand-based, as they rely on the prior knowledge of known actives to design similar molecules that potentially bind to the same target with the possibility to fine-tune their physicochemical properties 16,17 . To evaluate how well the new molecules bind to the target protein and prioritize them for synthesis, they can be followed up with subsequent structure-based docking calculations or molecular dynamics (MD) simulations 18 . The separation of molecular generation and structure-based evaluation, however, will render the whole process inefficient and severely limits the utility of molecular generative models in practical drug discovery projects.…”

Section: Introductionmentioning

confidence: 99%

A Simple Way to Incorporate Target Structural Information in Molecular Generative Models

Zhang

Huang

2023

Preprint

View full text Add to dashboard Cite

Deep learning generative models are now being applied in various fields including drug discovery. In this work, we propose a novel approach to include target 3D structural information in molecular generative models for structure-based drug design. The method combines a message-passing neural network model that predicts docking scores with a generative neural network model as its reward function to navigate the chemical space searching for molecules that bind favorably with a specific target. A key component of the method is to construct target-specific molecular sets for training to overcome potential transferability issues. Tests on eight target proteins showed a 100-fold increase in hit generation compared to conventional docking calculations, and the ability to generate molecules similar to approved drugs or known active ligands for specific targets without prior knowledge. This method provides a general and highly efficient solution for structure-based molecular generation.

show abstract

Section: Introductionmentioning

confidence: 99%

A Simple Way to Incorporate Target Structural Information in Molecular Generative Models

Zhang

Huang

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…[15,16] Molecular docking and dynamics are powerful approaches for understanding the structural interactions between binding molecules and protein targets, predicting the optimal binding form. [17,18] Furthermore, absorption, distribution, metabolism, excretion, and toxicity (ADME/Tox) analysis can be used to evaluate the bioavailability and toxicity profile of molecules. [19,20] In this present study, we employed Gaussian field-based 3D-QSAR and atom-based 3D-QSAR with the PHASE module (2021 version, Schrödinger, LLC, NY) to analyze a series of triazole derivatives with cytotoxic activities.…”

Section: Introductionmentioning

confidence: 99%

Computational Exploration of the Structural Requirements of Triazole Derivatives as Colchicine Binding Site Inhibitors

et al. 2023

View full text Add to dashboard Cite

Colchicine inhibits microtubule assembly by preventing tubulin polymerization, making the colchicine binding site a promising target against cancer. The present study focuses on the 3D‐QSAR analysis of 31 triazole analogs. The atom‐based and field‐based QSAR models exhibit significant statistical results for internal and external validations, with the atom‐based 3D‐QSAR model showing a Q2 of 0.658 and R2 of 0.998, and the field‐based 3D‐QSAR model showing a Q2 of 0.658 and R2 of 0.998. The 3D‐QSAR atom‐based pharmacophore model explains biological activity using a six‐point hypothesis (AHHRRR.1), while the field‐based 3D‐QSAR model explains activity based on steric, electrostatic, hydrophobic, and hydrogen bond donor/acceptor fields. Based on the 3D contour maps generated, five new compounds were designed and their ADME/Tox properties predicted. The results suggested that these compounds possess potent pharmacological characteristics. The results of the molecular docking analysis showed that Pred01 and Pred02 had higher docking scores (−6.2417Kcal/mol and −6.5049Kcal/mol, respectively) than compound 15, and they formed notable hydrogen bonds and hydrophobic interactions with active site residues. The MMGBSA results confirmed these findings and emphasized the essential role of van der Waals energy in the binding energy. Moreover, the dynamic stability of the complexes was further supported by MD simulations.

show abstract

“…Screening of virtual libraries of synthesizable compounds has become an increasingly important step in drug discovery 1 . The surge in utilization of computational approaches has been stimulated by improvements in binding energy calculations, the growth of computational resources, advances in protein structures determination and availability of large and diverse virtual libraries of compounds [2][3][4][5][6][7][8][9] . However, our ability to access the vast druggable chemical space is still limited and will be impacted by the availability of sufficient computing resources for the foreseeable future 10,11 .…”

Section: Introductionmentioning

confidence: 99%

Correlation of binding site properties with chemistries used for generation of ultra-large virtual libraries.

Song

Nicklaus

Tarasova

2023

Preprint

View full text Add to dashboard Cite

Although the size of virtual libraries of synthesizable compounds is growing rapidly, we are still enumerating only tiny fractions of the drug-like chemical universe. At the same time, our ability to mine these newly generated libraries also lags their growth. That is why fragment-based approaches that utilize on-demand virtual combinatorial libraries are gaining popularity. These à la carte libraries utilize synthetic blocks that have been shown to be effective binders in parts of target protein pockets. There is, however, no data on the potential impact of the chemistries used for making on-demand libraries on the hit rates during virtual screening. There are also no rules to guide in selection of these synthetic methods for libraries production. We have used the SAVI (Synthetically Accessible Virtual Inventory) library, constructed using 53 reliable reaction types (transforms), to test for correlations between these chemistries and docking hit rates for 39 well- characterized protein pockets. The data shows that the hit rate depends on the chemistry used and that chemistry selection can be optimized based on pocket properties.

show abstract

Docking-based generative approaches in the search for new drug candidates

Cited by 27 publications

References 58 publications

A Simple Way to Incorporate Target Structural Information in Molecular Generative Models

A Simple Way to Incorporate Target Structural Information in Molecular Generative Models

Computational Exploration of the Structural Requirements of Triazole Derivatives as Colchicine Binding Site Inhibitors

Correlation of binding site properties with chemistries used for generation of ultra-large virtual libraries.

Contact Info

Product

Resources

About