Deep Generative Models for 3D Linker Design
Rational compound design remains a challenging problem for both computational methods and medicinal chemists. Computational generative methods have begun to show promising results for the design problem. However, they have not yet used the power of three-dimensional (3D) structural information. We h...
Gespeichert in:
Veröffentlicht in: | Journal of chemical information and modeling 2020-04, Vol.60 (4), p.1983-1995 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 1995 |
---|---|
container_issue | 4 |
container_start_page | 1983 |
container_title | Journal of chemical information and modeling |
container_volume | 60 |
creator | Imrie, Fergus Bradley, Anthony R van der Schaar, Mihaela Deane, Charlotte M |
description | Rational compound design remains a challenging problem for both computational methods and medicinal chemists. Computational generative methods have begun to show promising results for the design problem. However, they have not yet used the power of three-dimensional (3D) structural information. We have developed a novel graph-based deep generative model that combines state-of-the-art machine learning techniques with structural knowledge. Our method (“DeLinker”) takes two fragments or partial structures and designs a molecule incorporating both. The generation process is protein-context-dependent, utilizing the relative distance and orientation between the partial structures. This 3D information is vital to successful compound design, and we demonstrate its impact on the generation process and the limitations of omitting such information. In a large-scale evaluation, DeLinker designed 60% more molecules with high 3D similarity to the original molecule than a database baseline. When considering the more relevant problem of longer linkers with at least five atoms, the outperformance increased to 200%. We demonstrate the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design. As far as we are aware, this is the first molecular generative model to incorporate 3D structural information directly in the design process. The code is available at https://github.com/oxpig/DeLinker. |
doi_str_mv | 10.1021/acs.jcim.9b01120 |
format | Article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7189367</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2381623668</sourcerecordid><originalsourceid>FETCH-LOGICAL-a461t-d9313366e195598ec20c8be1b3452692085427e14a6d409e151a9eb8f1bef19b3</originalsourceid><addsrcrecordid>eNp1kc1LxDAQxYMorq7ePUnBi6BdM0mbNhdBdnUVVrwoeAtpO1279mNN2gX_e7PuByp4msD83puZPEJOgA6AMrjSqR3M0qIayIQCMLpDDiAMpC8Ffd3dvEMpeuTQ2hmlnEvB9kmPM5BhGEcH5GKEOPfGWKPRbbFA77HJsLRe3hiPj7xJUb-j8UZoi2l9RPZyXVo8Xtc-ebm7fR7e-5On8cPwZuLrQEDrZ5ID50LgcoaMMWU0jROEhAchE5LROAxYhBBokQVUIoSgJSZxDgnmIBPeJ9cr33mXVJilWLdGl2puikqbT9XoQv3u1MWbmjYLFUEsuYicwfnawDQfHdpWVYVNsSx1jU1nFeMxCOZWjB169gedNZ2p3XmOkpEQkkvmKLqiUtNYazDfLgNULZNQLgm1TEKtk3CS059HbAWbr3fA5Qr4lm6G_uv3BcOKknU</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2397669392</pqid></control><display><type>article</type><title>Deep Generative Models for 3D Linker Design</title><source>MEDLINE</source><source>American Chemical Society Journals</source><creator>Imrie, Fergus ; Bradley, Anthony R ; van der Schaar, Mihaela ; Deane, Charlotte M</creator><creatorcontrib>Imrie, Fergus ; Bradley, Anthony R ; van der Schaar, Mihaela ; Deane, Charlotte M</creatorcontrib><description>Rational compound design remains a challenging problem for both computational methods and medicinal chemists. Computational generative methods have begun to show promising results for the design problem. However, they have not yet used the power of three-dimensional (3D) structural information. We have developed a novel graph-based deep generative model that combines state-of-the-art machine learning techniques with structural knowledge. Our method (“DeLinker”) takes two fragments or partial structures and designs a molecule incorporating both. The generation process is protein-context-dependent, utilizing the relative distance and orientation between the partial structures. This 3D information is vital to successful compound design, and we demonstrate its impact on the generation process and the limitations of omitting such information. In a large-scale evaluation, DeLinker designed 60% more molecules with high 3D similarity to the original molecule than a database baseline. When considering the more relevant problem of longer linkers with at least five atoms, the outperformance increased to 200%. We demonstrate the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design. As far as we are aware, this is the first molecular generative model to incorporate 3D structural information directly in the design process. The code is available at https://github.com/oxpig/DeLinker.</description><identifier>ISSN: 1549-9596</identifier><identifier>EISSN: 1549-960X</identifier><identifier>DOI: 10.1021/acs.jcim.9b01120</identifier><identifier>PMID: 32195587</identifier><language>eng</language><publisher>United States: American Chemical Society</publisher><subject>Algorithms ; Chemists ; Design ; Machine Learning ; Models, Molecular ; Proteins ; Three dimensional models</subject><ispartof>Journal of chemical information and modeling, 2020-04, Vol.60 (4), p.1983-1995</ispartof><rights>Copyright American Chemical Society Apr 27, 2020</rights><rights>Copyright © 2020 American Chemical Society 2020 American Chemical Society</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-a461t-d9313366e195598ec20c8be1b3452692085427e14a6d409e151a9eb8f1bef19b3</citedby><cites>FETCH-LOGICAL-a461t-d9313366e195598ec20c8be1b3452692085427e14a6d409e151a9eb8f1bef19b3</cites><orcidid>0000-0003-1388-2252 ; 0000-0002-6241-0123</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://pubs.acs.org/doi/pdf/10.1021/acs.jcim.9b01120$$EPDF$$P50$$Gacs$$H</linktopdf><linktohtml>$$Uhttps://pubs.acs.org/doi/10.1021/acs.jcim.9b01120$$EHTML$$P50$$Gacs$$H</linktohtml><link.rule.ids>230,314,780,784,885,2765,27076,27924,27925,56738,56788</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/32195587$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Imrie, Fergus</creatorcontrib><creatorcontrib>Bradley, Anthony R</creatorcontrib><creatorcontrib>van der Schaar, Mihaela</creatorcontrib><creatorcontrib>Deane, Charlotte M</creatorcontrib><title>Deep Generative Models for 3D Linker Design</title><title>Journal of chemical information and modeling</title><addtitle>J. Chem. Inf. Model</addtitle><description>Rational compound design remains a challenging problem for both computational methods and medicinal chemists. Computational generative methods have begun to show promising results for the design problem. However, they have not yet used the power of three-dimensional (3D) structural information. We have developed a novel graph-based deep generative model that combines state-of-the-art machine learning techniques with structural knowledge. Our method (“DeLinker”) takes two fragments or partial structures and designs a molecule incorporating both. The generation process is protein-context-dependent, utilizing the relative distance and orientation between the partial structures. This 3D information is vital to successful compound design, and we demonstrate its impact on the generation process and the limitations of omitting such information. In a large-scale evaluation, DeLinker designed 60% more molecules with high 3D similarity to the original molecule than a database baseline. When considering the more relevant problem of longer linkers with at least five atoms, the outperformance increased to 200%. We demonstrate the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design. As far as we are aware, this is the first molecular generative model to incorporate 3D structural information directly in the design process. The code is available at https://github.com/oxpig/DeLinker.</description><subject>Algorithms</subject><subject>Chemists</subject><subject>Design</subject><subject>Machine Learning</subject><subject>Models, Molecular</subject><subject>Proteins</subject><subject>Three dimensional models</subject><issn>1549-9596</issn><issn>1549-960X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp1kc1LxDAQxYMorq7ePUnBi6BdM0mbNhdBdnUVVrwoeAtpO1279mNN2gX_e7PuByp4msD83puZPEJOgA6AMrjSqR3M0qIayIQCMLpDDiAMpC8Ffd3dvEMpeuTQ2hmlnEvB9kmPM5BhGEcH5GKEOPfGWKPRbbFA77HJsLRe3hiPj7xJUb-j8UZoi2l9RPZyXVo8Xtc-ebm7fR7e-5On8cPwZuLrQEDrZ5ID50LgcoaMMWU0jROEhAchE5LROAxYhBBokQVUIoSgJSZxDgnmIBPeJ9cr33mXVJilWLdGl2puikqbT9XoQv3u1MWbmjYLFUEsuYicwfnawDQfHdpWVYVNsSx1jU1nFeMxCOZWjB169gedNZ2p3XmOkpEQkkvmKLqiUtNYazDfLgNULZNQLgm1TEKtk3CS059HbAWbr3fA5Qr4lm6G_uv3BcOKknU</recordid><startdate>20200427</startdate><enddate>20200427</enddate><creator>Imrie, Fergus</creator><creator>Bradley, Anthony R</creator><creator>van der Schaar, Mihaela</creator><creator>Deane, Charlotte M</creator><general>American Chemical Society</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SR</scope><scope>7U5</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0003-1388-2252</orcidid><orcidid>https://orcid.org/0000-0002-6241-0123</orcidid></search><sort><creationdate>20200427</creationdate><title>Deep Generative Models for 3D Linker Design</title><author>Imrie, Fergus ; Bradley, Anthony R ; van der Schaar, Mihaela ; Deane, Charlotte M</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a461t-d9313366e195598ec20c8be1b3452692085427e14a6d409e151a9eb8f1bef19b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Algorithms</topic><topic>Chemists</topic><topic>Design</topic><topic>Machine Learning</topic><topic>Models, Molecular</topic><topic>Proteins</topic><topic>Three dimensional models</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Imrie, Fergus</creatorcontrib><creatorcontrib>Bradley, Anthony R</creatorcontrib><creatorcontrib>van der Schaar, Mihaela</creatorcontrib><creatorcontrib>Deane, Charlotte M</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of chemical information and modeling</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Imrie, Fergus</au><au>Bradley, Anthony R</au><au>van der Schaar, Mihaela</au><au>Deane, Charlotte M</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Deep Generative Models for 3D Linker Design</atitle><jtitle>Journal of chemical information and modeling</jtitle><addtitle>J. Chem. Inf. Model</addtitle><date>2020-04-27</date><risdate>2020</risdate><volume>60</volume><issue>4</issue><spage>1983</spage><epage>1995</epage><pages>1983-1995</pages><issn>1549-9596</issn><eissn>1549-960X</eissn><abstract>Rational compound design remains a challenging problem for both computational methods and medicinal chemists. Computational generative methods have begun to show promising results for the design problem. However, they have not yet used the power of three-dimensional (3D) structural information. We have developed a novel graph-based deep generative model that combines state-of-the-art machine learning techniques with structural knowledge. Our method (“DeLinker”) takes two fragments or partial structures and designs a molecule incorporating both. The generation process is protein-context-dependent, utilizing the relative distance and orientation between the partial structures. This 3D information is vital to successful compound design, and we demonstrate its impact on the generation process and the limitations of omitting such information. In a large-scale evaluation, DeLinker designed 60% more molecules with high 3D similarity to the original molecule than a database baseline. When considering the more relevant problem of longer linkers with at least five atoms, the outperformance increased to 200%. We demonstrate the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design. As far as we are aware, this is the first molecular generative model to incorporate 3D structural information directly in the design process. The code is available at https://github.com/oxpig/DeLinker.</abstract><cop>United States</cop><pub>American Chemical Society</pub><pmid>32195587</pmid><doi>10.1021/acs.jcim.9b01120</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0003-1388-2252</orcidid><orcidid>https://orcid.org/0000-0002-6241-0123</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1549-9596 |
ispartof | Journal of chemical information and modeling, 2020-04, Vol.60 (4), p.1983-1995 |
issn | 1549-9596 1549-960X |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7189367 |
source | MEDLINE; American Chemical Society Journals |
subjects | Algorithms Chemists Design Machine Learning Models, Molecular Proteins Three dimensional models |
title | Deep Generative Models for 3D Linker Design |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T16%3A33%3A25IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Deep%20Generative%20Models%20for%203D%20Linker%20Design&rft.jtitle=Journal%20of%20chemical%20information%20and%20modeling&rft.au=Imrie,%20Fergus&rft.date=2020-04-27&rft.volume=60&rft.issue=4&rft.spage=1983&rft.epage=1995&rft.pages=1983-1995&rft.issn=1549-9596&rft.eissn=1549-960X&rft_id=info:doi/10.1021/acs.jcim.9b01120&rft_dat=%3Cproquest_pubme%3E2381623668%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2397669392&rft_id=info:pmid/32195587&rfr_iscdi=true |