V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization

We propose a computational workflow to design novel drug-like molecules by combining the global optimization of molecular properties and protein-ligand docking with machine learning. However, most existing methods depend heavily on experimental data, and many targets do not have sufficient data to t...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Jieun Choi, Juyong Lee
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/a7adeb7431c840a1aff8c538162a106c
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:a7adeb7431c840a1aff8c538162a106c
record_format dspace
spelling oai:doaj.org-article:a7adeb7431c840a1aff8c538162a106c2021-11-11T17:06:38ZV-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization10.3390/ijms2221116351422-00671661-6596https://doaj.org/article/a7adeb7431c840a1aff8c538162a106c2021-10-01T00:00:00Zhttps://www.mdpi.com/1422-0067/22/21/11635https://doaj.org/toc/1661-6596https://doaj.org/toc/1422-0067We propose a computational workflow to design novel drug-like molecules by combining the global optimization of molecular properties and protein-ligand docking with machine learning. However, most existing methods depend heavily on experimental data, and many targets do not have sufficient data to train reliable activity prediction models. To overcome this limitation, protein-ligand docking calculations must be performed using the limited data available. Such docking calculations during molecular generation require considerable computational time, preventing extensive exploration of the chemical space. To address this problem, we trained a machine-learning-based model that predicted the docking energy using SMILES to accelerate the molecular generation process. Docking scores could be accurately predicted using only a SMILES string. We combined this docking score prediction model with the global molecular property optimization approach, MolFinder, to find novel molecules exhibiting the desired properties with high values of predicted docking scores. We named this design approach V-dock. Using V-dock, we efficiently generated many novel molecules with high docking scores for a target protein, a similarity to the reference molecule, and desirable drug-like and bespoke properties, such as QED. The predicted docking scores of the generated molecules were verified by correlating them with the actual docking scores.Jieun ChoiJuyong LeeMDPI AGarticleprotein-ligand dockingcomputer-aided drug discoverydocking score predictionquantitative estimation of drug-likeness (QED)conformational space annealing (CSA)Lipinski’s rule of fiveBiology (General)QH301-705.5ChemistryQD1-999ENInternational Journal of Molecular Sciences, Vol 22, Iss 11635, p 11635 (2021)
institution DOAJ
collection DOAJ
language EN
topic protein-ligand docking
computer-aided drug discovery
docking score prediction
quantitative estimation of drug-likeness (QED)
conformational space annealing (CSA)
Lipinski’s rule of five
Biology (General)
QH301-705.5
Chemistry
QD1-999
spellingShingle protein-ligand docking
computer-aided drug discovery
docking score prediction
quantitative estimation of drug-likeness (QED)
conformational space annealing (CSA)
Lipinski’s rule of five
Biology (General)
QH301-705.5
Chemistry
QD1-999
Jieun Choi
Juyong Lee
V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
description We propose a computational workflow to design novel drug-like molecules by combining the global optimization of molecular properties and protein-ligand docking with machine learning. However, most existing methods depend heavily on experimental data, and many targets do not have sufficient data to train reliable activity prediction models. To overcome this limitation, protein-ligand docking calculations must be performed using the limited data available. Such docking calculations during molecular generation require considerable computational time, preventing extensive exploration of the chemical space. To address this problem, we trained a machine-learning-based model that predicted the docking energy using SMILES to accelerate the molecular generation process. Docking scores could be accurately predicted using only a SMILES string. We combined this docking score prediction model with the global molecular property optimization approach, MolFinder, to find novel molecules exhibiting the desired properties with high values of predicted docking scores. We named this design approach V-dock. Using V-dock, we efficiently generated many novel molecules with high docking scores for a target protein, a similarity to the reference molecule, and desirable drug-like and bespoke properties, such as QED. The predicted docking scores of the generated molecules were verified by correlating them with the actual docking scores.
format article
author Jieun Choi
Juyong Lee
author_facet Jieun Choi
Juyong Lee
author_sort Jieun Choi
title V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
title_short V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
title_full V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
title_fullStr V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
title_full_unstemmed V-Dock: Fast Generation of Novel Drug-like Molecules Using Machine-Learning-Based Docking Score and Molecular Optimization
title_sort v-dock: fast generation of novel drug-like molecules using machine-learning-based docking score and molecular optimization
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/a7adeb7431c840a1aff8c538162a106c
work_keys_str_mv AT jieunchoi vdockfastgenerationofnoveldruglikemoleculesusingmachinelearningbaseddockingscoreandmolecularoptimization
AT juyonglee vdockfastgenerationofnoveldruglikemoleculesusingmachinelearningbaseddockingscoreandmolecularoptimization
_version_ 1718432195203301376