Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques

The cytochrome P450 (P450) superfamily plays an important role in the metabolism of drug compounds, and it is therefore highly desirable to have models that can predict whether a compound interacts with a specific isoform of the P450s. In this work, we provide in silico models for classification of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Drug metabolism and disposition 2009-03, Vol.37 (3), p.658-664
Hauptverfasser: Vasanthanathan, Poongavanam, Taboureau, Olivier, Oostenbrink, Chris, Vermeulen, Nico P.E., Olsen, Lars, Jørgensen, Flemming Steen
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 664
container_issue 3
container_start_page 658
container_title Drug metabolism and disposition
container_volume 37
creator Vasanthanathan, Poongavanam
Taboureau, Olivier
Oostenbrink, Chris
Vermeulen, Nico P.E.
Olsen, Lars
Jørgensen, Flemming Steen
description The cytochrome P450 (P450) superfamily plays an important role in the metabolism of drug compounds, and it is therefore highly desirable to have models that can predict whether a compound interacts with a specific isoform of the P450s. In this work, we provide in silico models for classification of CYP1A2 inhibitors and noninhibitors. Training and test sets consisted of approximately 400 and 7000 compounds, respectively. Various machine learning techniques, such as binary quantitative structure activity relationship, support vector machine (SVM), random forest, kappa nearest neighbor (kNN), and decision tree methods were used to develop in silico models, based on Volsurf and Molecular Operating Environment descriptors. The best models were obtained using the SVM, random forest, and kNN methods in combination with the BestFirst variable selection method, resulting in models with 73 to 76% of accuracy on the test set prediction (Matthews correlation coefficients of 0.51 and 0.52). Finally, a decision tree model based on Lipinski’s Rule-of-Five descriptors was also developed. This model predicts 67% of the compounds correctly and gives a simple and interesting insight into the issue of classification. All of the models developed in this work are fast and precise enough to be applicable for virtual screening of CYP1A2 inhibitors or noninhibitors or can be used as simple filters in the drug discovery process.
doi_str_mv 10.1124/dmd.108.023507
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_66940988</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0090955624019883</els_id><sourcerecordid>66940988</sourcerecordid><originalsourceid>FETCH-LOGICAL-c468t-5ca8d3c175168fbf67b314321a998b2b252b13f5b26c5ea4c4365c6c522cf7553</originalsourceid><addsrcrecordid>eNp1kEuP0zAQgC0EYsvClSPyBW4pHjt2nOOqWmCl8jgsEjfLmTiNUWIXOwX13-OqFT1xmtHMNw99hLwGtgbg9ft-7tfA9JpxIVnzhKxAcqgYa388JasSWNVKqW7Ii5x_MgZ1Ldrn5AZaJlULckXsZrI5-8GjXXwMNA50c1wijinOjn6rJaNwx-lDGH3nl5gytaGnX2Lw10p3pJ8tjj44unU2ldaOPjocg_91cPkleTbYKbtXl3hLvn-4f9x8qrZfPz5s7rYV1kovlUSre4HQSFB66AbVdAJqwcG2re54xyXvQAyy4wqlszXWQkksOec4NFKKW_LuvHef4unuYmaf0U2TDS4eslGqrVmrdQHXZxBTzDm5weyTn206GmDmJNUUqSXX5iy1DLy5bD50s-uv-MViAd5eAJvRTkOyAX3-x3EArRvQV270u_GPT87sR5tmi3GKu6MRjRFGyROnz5wrvn57l0xG7wK6vszgYvro__frX2ZDnqM</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>66940988</pqid></control><display><type>article</type><title>Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques</title><source>MEDLINE</source><source>Alma/SFX Local Collection</source><creator>Vasanthanathan, Poongavanam ; Taboureau, Olivier ; Oostenbrink, Chris ; Vermeulen, Nico P.E. ; Olsen, Lars ; Jørgensen, Flemming Steen</creator><creatorcontrib>Vasanthanathan, Poongavanam ; Taboureau, Olivier ; Oostenbrink, Chris ; Vermeulen, Nico P.E. ; Olsen, Lars ; Jørgensen, Flemming Steen</creatorcontrib><description>The cytochrome P450 (P450) superfamily plays an important role in the metabolism of drug compounds, and it is therefore highly desirable to have models that can predict whether a compound interacts with a specific isoform of the P450s. In this work, we provide in silico models for classification of CYP1A2 inhibitors and noninhibitors. Training and test sets consisted of approximately 400 and 7000 compounds, respectively. Various machine learning techniques, such as binary quantitative structure activity relationship, support vector machine (SVM), random forest, kappa nearest neighbor (kNN), and decision tree methods were used to develop in silico models, based on Volsurf and Molecular Operating Environment descriptors. The best models were obtained using the SVM, random forest, and kNN methods in combination with the BestFirst variable selection method, resulting in models with 73 to 76% of accuracy on the test set prediction (Matthews correlation coefficients of 0.51 and 0.52). Finally, a decision tree model based on Lipinski’s Rule-of-Five descriptors was also developed. This model predicts 67% of the compounds correctly and gives a simple and interesting insight into the issue of classification. All of the models developed in this work are fast and precise enough to be applicable for virtual screening of CYP1A2 inhibitors or noninhibitors or can be used as simple filters in the drug discovery process.</description><identifier>ISSN: 0090-9556</identifier><identifier>EISSN: 1521-009X</identifier><identifier>DOI: 10.1124/dmd.108.023507</identifier><identifier>PMID: 19056915</identifier><identifier>CODEN: DMDSAI</identifier><language>eng</language><publisher>Bethesda, MD: Elsevier Inc</publisher><subject>Artificial Intelligence ; Biological and medical sciences ; Cytochrome P-450 CYP1A2 Inhibitors ; Enzyme Inhibitors - classification ; Medical sciences ; Models, Theoretical ; Pharmacology. Drug treatments ; Quantitative Structure-Activity Relationship</subject><ispartof>Drug metabolism and disposition, 2009-03, Vol.37 (3), p.658-664</ispartof><rights>2009 American Society for Pharmacology and Experimental Therapeutics</rights><rights>2009 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c468t-5ca8d3c175168fbf67b314321a998b2b252b13f5b26c5ea4c4365c6c522cf7553</citedby><cites>FETCH-LOGICAL-c468t-5ca8d3c175168fbf67b314321a998b2b252b13f5b26c5ea4c4365c6c522cf7553</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,27901,27902</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&amp;idt=21188718$$DView record in Pascal Francis$$Hfree_for_read</backlink><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/19056915$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Vasanthanathan, Poongavanam</creatorcontrib><creatorcontrib>Taboureau, Olivier</creatorcontrib><creatorcontrib>Oostenbrink, Chris</creatorcontrib><creatorcontrib>Vermeulen, Nico P.E.</creatorcontrib><creatorcontrib>Olsen, Lars</creatorcontrib><creatorcontrib>Jørgensen, Flemming Steen</creatorcontrib><title>Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques</title><title>Drug metabolism and disposition</title><addtitle>Drug Metab Dispos</addtitle><description>The cytochrome P450 (P450) superfamily plays an important role in the metabolism of drug compounds, and it is therefore highly desirable to have models that can predict whether a compound interacts with a specific isoform of the P450s. In this work, we provide in silico models for classification of CYP1A2 inhibitors and noninhibitors. Training and test sets consisted of approximately 400 and 7000 compounds, respectively. Various machine learning techniques, such as binary quantitative structure activity relationship, support vector machine (SVM), random forest, kappa nearest neighbor (kNN), and decision tree methods were used to develop in silico models, based on Volsurf and Molecular Operating Environment descriptors. The best models were obtained using the SVM, random forest, and kNN methods in combination with the BestFirst variable selection method, resulting in models with 73 to 76% of accuracy on the test set prediction (Matthews correlation coefficients of 0.51 and 0.52). Finally, a decision tree model based on Lipinski’s Rule-of-Five descriptors was also developed. This model predicts 67% of the compounds correctly and gives a simple and interesting insight into the issue of classification. All of the models developed in this work are fast and precise enough to be applicable for virtual screening of CYP1A2 inhibitors or noninhibitors or can be used as simple filters in the drug discovery process.</description><subject>Artificial Intelligence</subject><subject>Biological and medical sciences</subject><subject>Cytochrome P-450 CYP1A2 Inhibitors</subject><subject>Enzyme Inhibitors - classification</subject><subject>Medical sciences</subject><subject>Models, Theoretical</subject><subject>Pharmacology. Drug treatments</subject><subject>Quantitative Structure-Activity Relationship</subject><issn>0090-9556</issn><issn>1521-009X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2009</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp1kEuP0zAQgC0EYsvClSPyBW4pHjt2nOOqWmCl8jgsEjfLmTiNUWIXOwX13-OqFT1xmtHMNw99hLwGtgbg9ft-7tfA9JpxIVnzhKxAcqgYa388JasSWNVKqW7Ii5x_MgZ1Ldrn5AZaJlULckXsZrI5-8GjXXwMNA50c1wijinOjn6rJaNwx-lDGH3nl5gytaGnX2Lw10p3pJ8tjj44unU2ldaOPjocg_91cPkleTbYKbtXl3hLvn-4f9x8qrZfPz5s7rYV1kovlUSre4HQSFB66AbVdAJqwcG2re54xyXvQAyy4wqlszXWQkksOec4NFKKW_LuvHef4unuYmaf0U2TDS4eslGqrVmrdQHXZxBTzDm5weyTn206GmDmJNUUqSXX5iy1DLy5bD50s-uv-MViAd5eAJvRTkOyAX3-x3EArRvQV270u_GPT87sR5tmi3GKu6MRjRFGyROnz5wrvn57l0xG7wK6vszgYvro__frX2ZDnqM</recordid><startdate>20090301</startdate><enddate>20090301</enddate><creator>Vasanthanathan, Poongavanam</creator><creator>Taboureau, Olivier</creator><creator>Oostenbrink, Chris</creator><creator>Vermeulen, Nico P.E.</creator><creator>Olsen, Lars</creator><creator>Jørgensen, Flemming Steen</creator><general>Elsevier Inc</general><general>American Society for Pharmacology and Experimental Therapeutics</general><scope>IQODW</scope><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>7X8</scope></search><sort><creationdate>20090301</creationdate><title>Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques</title><author>Vasanthanathan, Poongavanam ; Taboureau, Olivier ; Oostenbrink, Chris ; Vermeulen, Nico P.E. ; Olsen, Lars ; Jørgensen, Flemming Steen</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c468t-5ca8d3c175168fbf67b314321a998b2b252b13f5b26c5ea4c4365c6c522cf7553</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2009</creationdate><topic>Artificial Intelligence</topic><topic>Biological and medical sciences</topic><topic>Cytochrome P-450 CYP1A2 Inhibitors</topic><topic>Enzyme Inhibitors - classification</topic><topic>Medical sciences</topic><topic>Models, Theoretical</topic><topic>Pharmacology. Drug treatments</topic><topic>Quantitative Structure-Activity Relationship</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Vasanthanathan, Poongavanam</creatorcontrib><creatorcontrib>Taboureau, Olivier</creatorcontrib><creatorcontrib>Oostenbrink, Chris</creatorcontrib><creatorcontrib>Vermeulen, Nico P.E.</creatorcontrib><creatorcontrib>Olsen, Lars</creatorcontrib><creatorcontrib>Jørgensen, Flemming Steen</creatorcontrib><collection>Pascal-Francis</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Drug metabolism and disposition</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Vasanthanathan, Poongavanam</au><au>Taboureau, Olivier</au><au>Oostenbrink, Chris</au><au>Vermeulen, Nico P.E.</au><au>Olsen, Lars</au><au>Jørgensen, Flemming Steen</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques</atitle><jtitle>Drug metabolism and disposition</jtitle><addtitle>Drug Metab Dispos</addtitle><date>2009-03-01</date><risdate>2009</risdate><volume>37</volume><issue>3</issue><spage>658</spage><epage>664</epage><pages>658-664</pages><issn>0090-9556</issn><eissn>1521-009X</eissn><coden>DMDSAI</coden><abstract>The cytochrome P450 (P450) superfamily plays an important role in the metabolism of drug compounds, and it is therefore highly desirable to have models that can predict whether a compound interacts with a specific isoform of the P450s. In this work, we provide in silico models for classification of CYP1A2 inhibitors and noninhibitors. Training and test sets consisted of approximately 400 and 7000 compounds, respectively. Various machine learning techniques, such as binary quantitative structure activity relationship, support vector machine (SVM), random forest, kappa nearest neighbor (kNN), and decision tree methods were used to develop in silico models, based on Volsurf and Molecular Operating Environment descriptors. The best models were obtained using the SVM, random forest, and kNN methods in combination with the BestFirst variable selection method, resulting in models with 73 to 76% of accuracy on the test set prediction (Matthews correlation coefficients of 0.51 and 0.52). Finally, a decision tree model based on Lipinski’s Rule-of-Five descriptors was also developed. This model predicts 67% of the compounds correctly and gives a simple and interesting insight into the issue of classification. All of the models developed in this work are fast and precise enough to be applicable for virtual screening of CYP1A2 inhibitors or noninhibitors or can be used as simple filters in the drug discovery process.</abstract><cop>Bethesda, MD</cop><pub>Elsevier Inc</pub><pmid>19056915</pmid><doi>10.1124/dmd.108.023507</doi><tpages>7</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0090-9556
ispartof Drug metabolism and disposition, 2009-03, Vol.37 (3), p.658-664
issn 0090-9556
1521-009X
language eng
recordid cdi_proquest_miscellaneous_66940988
source MEDLINE; Alma/SFX Local Collection
subjects Artificial Intelligence
Biological and medical sciences
Cytochrome P-450 CYP1A2 Inhibitors
Enzyme Inhibitors - classification
Medical sciences
Models, Theoretical
Pharmacology. Drug treatments
Quantitative Structure-Activity Relationship
title Classification of Cytochrome P450 1A2 Inhibitors and Noninhibitors by Machine Learning Techniques
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T06%3A05%3A47IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Classification%20of%20Cytochrome%20P450%201A2%20Inhibitors%20and%20Noninhibitors%20by%20Machine%20Learning%20Techniques&rft.jtitle=Drug%20metabolism%20and%20disposition&rft.au=Vasanthanathan,%20Poongavanam&rft.date=2009-03-01&rft.volume=37&rft.issue=3&rft.spage=658&rft.epage=664&rft.pages=658-664&rft.issn=0090-9556&rft.eissn=1521-009X&rft.coden=DMDSAI&rft_id=info:doi/10.1124/dmd.108.023507&rft_dat=%3Cproquest_cross%3E66940988%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=66940988&rft_id=info:pmid/19056915&rft_els_id=S0090955624019883&rfr_iscdi=true