Active search for computer-aided drug design

We consider lead discovery as active search in a space of labelled graphs. In particular, we extend our recent data-driven adaptive Markov chain approach, and evaluate it on a focused drug design problem, where we search for an antagonist of an av integrin, the target protein that belongs to a group...

Full description

Bibliographic Details
Main Authors: Oglic, Dino, Oatley, Steven A., Macdonald, Simon J.F., McInally, Thomas, Garnett, Roman, Hirst, Jonathan D., Gärtner, Thomas
Format: Article
Published: Wiley-VCH Verlag 2018
Subjects:
Online Access:https://eprints.nottingham.ac.uk/49505/
http://dx.doi.org/10.1002/minf.201700130
_version_ 1848798013266329600
author Oglic, Dino
Oatley, Steven A.
Macdonald, Simon J.F.
McInally, Thomas
Garnett, Roman
Hirst, Jonathan D.
Gärtner, Thomas
author_facet Oglic, Dino
Oatley, Steven A.
Macdonald, Simon J.F.
McInally, Thomas
Garnett, Roman
Hirst, Jonathan D.
Gärtner, Thomas
author_sort Oglic, Dino
building Nottingham Research Data Repository
collection Online Access
description We consider lead discovery as active search in a space of labelled graphs. In particular, we extend our recent data-driven adaptive Markov chain approach, and evaluate it on a focused drug design problem, where we search for an antagonist of an av integrin, the target protein that belongs to a group of Arg-Gly-Asp integrin receptors. This group of integrin receptors is thought to play a key role in idiopathic pulmonary fibrosis, a chronic lung disease of significant pharmaceutical interest. As an in silico proxy of the binding affinity, we use a molecular docking score to an experimentally determined avb6 protein structure. The search is driven by a probabilistic surrogate of the activity of all molecules from that space. As the process evolves and the algorithm observes the activity scores of the previously designed molecules, the hypothesis of the activity is refined. The algorithm is guaranteed to converge in probability to the best hypothesis from an a priori specified hypothesis space. In our empirical evaluations, the approach achieves a large structural variety of designed molecular structures for which the docking score is better than the desired threshold. Some novel molecules, suggested to be active by the surrogate model, provoke a significant interest from the perspective of medicinal chemistry and warrant prioritization for synthesis. Moreover, the approach discovered 19 out of the 24 active compounds which are known to be active from previous biological assays.
first_indexed 2025-11-14T20:13:01Z
format Article
id nottingham-49505
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:13:01Z
publishDate 2018
publisher Wiley-VCH Verlag
recordtype eprints
repository_type Digital Repository
spelling nottingham-495052020-05-04T19:30:22Z https://eprints.nottingham.ac.uk/49505/ Active search for computer-aided drug design Oglic, Dino Oatley, Steven A. Macdonald, Simon J.F. McInally, Thomas Garnett, Roman Hirst, Jonathan D. Gärtner, Thomas We consider lead discovery as active search in a space of labelled graphs. In particular, we extend our recent data-driven adaptive Markov chain approach, and evaluate it on a focused drug design problem, where we search for an antagonist of an av integrin, the target protein that belongs to a group of Arg-Gly-Asp integrin receptors. This group of integrin receptors is thought to play a key role in idiopathic pulmonary fibrosis, a chronic lung disease of significant pharmaceutical interest. As an in silico proxy of the binding affinity, we use a molecular docking score to an experimentally determined avb6 protein structure. The search is driven by a probabilistic surrogate of the activity of all molecules from that space. As the process evolves and the algorithm observes the activity scores of the previously designed molecules, the hypothesis of the activity is refined. The algorithm is guaranteed to converge in probability to the best hypothesis from an a priori specified hypothesis space. In our empirical evaluations, the approach achieves a large structural variety of designed molecular structures for which the docking score is better than the desired threshold. Some novel molecules, suggested to be active by the surrogate model, provoke a significant interest from the perspective of medicinal chemistry and warrant prioritization for synthesis. Moreover, the approach discovered 19 out of the 24 active compounds which are known to be active from previous biological assays. Wiley-VCH Verlag 2018-02-01 Article PeerReviewed Oglic, Dino, Oatley, Steven A., Macdonald, Simon J.F., McInally, Thomas, Garnett, Roman, Hirst, Jonathan D. and Gärtner, Thomas (2018) Active search for computer-aided drug design. Molecular Informatics, 37 . p. 1700130. ISSN 1868-1743 active search ; antagonist ; cheminformatics ; drug design ; integrin http://onlinelibrary.wiley.com/doi/10.1002/minf.201700130/abstract http://dx.doi.org/10.1002/minf.201700130 http://dx.doi.org/10.1002/minf.201700130
spellingShingle active search ; antagonist ; cheminformatics ; drug design ; integrin
Oglic, Dino
Oatley, Steven A.
Macdonald, Simon J.F.
McInally, Thomas
Garnett, Roman
Hirst, Jonathan D.
Gärtner, Thomas
Active search for computer-aided drug design
title Active search for computer-aided drug design
title_full Active search for computer-aided drug design
title_fullStr Active search for computer-aided drug design
title_full_unstemmed Active search for computer-aided drug design
title_short Active search for computer-aided drug design
title_sort active search for computer-aided drug design
topic active search ; antagonist ; cheminformatics ; drug design ; integrin
url https://eprints.nottingham.ac.uk/49505/
https://eprints.nottingham.ac.uk/49505/
https://eprints.nottingham.ac.uk/49505/
http://dx.doi.org/10.1002/minf.201700130