Modelling Chemical Reasoning to Predict and Invent Reactions
The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a...
Gespeichert in:
Veröffentlicht in: | Chemistry : a European journal 2017-05, Vol.23 (25), p.6118-6128 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 6128 |
---|---|
container_issue | 25 |
container_start_page | 6118 |
container_title | Chemistry : a European journal |
container_volume | 23 |
creator | Segler, Marwin H. S. Waller, Mark P. |
description | The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph containing 14.4 million molecules and 8.2 million binary reactions, which represents the bulk of all chemical reactions ever published in the scientific literature. Our model outperforms a rule‐based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data‐driven model generalises even beyond known reaction types, and is thus capable of effectively (re‐)discovering novel transformations (even including transition metal‐catalysed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub‐second time frame, the model can be used as a high‐throughput generator of reaction hypotheses for reaction discovery.
Learn from past experience: The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations (see figure). A model has been developed to mimic chemical reasoning by formalising reaction prediction as finding missing links in a knowledge graph. |
doi_str_mv | 10.1002/chem.201604556 |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1841796490</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1841796490</sourcerecordid><originalsourceid>FETCH-LOGICAL-c4106-f20d1691e3dbc47d53bad0d922f4a8c785142f3865c0a86939783d0ae5863f923</originalsourceid><addsrcrecordid>eNqFkE1Lw0AQhhdRbK1ePUrAi5fU2e9d8CKhWqFFET2H7e5GU9KkZhOl_96E1gpePA3MPPPw8iJ0jmGMAci1fferMQEsgHEuDtAQc4JjKgU_REPQTMaCUz1AJyEsAUALSo_RgEglCJNyiG7mlfNFkZdvUdKpcmuK6NmbUJX9qqmip9q73DaRKV30UH76sunvtsmrMpyio8wUwZ_t5gi93k1ekmk8e7x_SG5nsWUYRJwRcFho7KlbWCYdpwvjwGlCMmaUlYpjRjKqBLdglNBUS0UdGM-VoJkmdISutt51XX20PjTpKg-2i21KX7UhxYphqQXT0KGXf9Bl1dZll66jNAfFCe2p8ZaydRVC7bN0XecrU29SDGnfa9r3mu577R4udtp2sfJuj_8U2QF6C3zlhd_8o0uT6WT-K_8G_pKB5A</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1895085230</pqid></control><display><type>article</type><title>Modelling Chemical Reasoning to Predict and Invent Reactions</title><source>Wiley Online Library - AutoHoldings Journals</source><creator>Segler, Marwin H. S. ; Waller, Mark P.</creator><creatorcontrib>Segler, Marwin H. S. ; Waller, Mark P.</creatorcontrib><description>The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph containing 14.4 million molecules and 8.2 million binary reactions, which represents the bulk of all chemical reactions ever published in the scientific literature. Our model outperforms a rule‐based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data‐driven model generalises even beyond known reaction types, and is thus capable of effectively (re‐)discovering novel transformations (even including transition metal‐catalysed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub‐second time frame, the model can be used as a high‐throughput generator of reaction hypotheses for reaction discovery.
Learn from past experience: The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations (see figure). A model has been developed to mimic chemical reasoning by formalising reaction prediction as finding missing links in a knowledge graph.</description><identifier>ISSN: 0947-6539</identifier><identifier>EISSN: 1521-3765</identifier><identifier>DOI: 10.1002/chem.201604556</identifier><identifier>PMID: 27862477</identifier><language>eng</language><publisher>Germany: Wiley Subscription Services, Inc</publisher><subject>artificial intelligence ; augmented scientific discovery ; Chemical reactions ; Chemistry ; Chemists ; computational chemistry ; Computers ; Expert systems ; graph theory ; Hypotheses ; Knowledge representation ; organic chemistry ; Predictions ; Reasoning ; Transition metals</subject><ispartof>Chemistry : a European journal, 2017-05, Vol.23 (25), p.6118-6128</ispartof><rights>2017 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim</rights><rights>2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.</rights><rights>2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c4106-f20d1691e3dbc47d53bad0d922f4a8c785142f3865c0a86939783d0ae5863f923</citedby><cites>FETCH-LOGICAL-c4106-f20d1691e3dbc47d53bad0d922f4a8c785142f3865c0a86939783d0ae5863f923</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1002%2Fchem.201604556$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1002%2Fchem.201604556$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>314,780,784,1417,27924,27925,45574,45575</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/27862477$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Segler, Marwin H. S.</creatorcontrib><creatorcontrib>Waller, Mark P.</creatorcontrib><title>Modelling Chemical Reasoning to Predict and Invent Reactions</title><title>Chemistry : a European journal</title><addtitle>Chemistry</addtitle><description>The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph containing 14.4 million molecules and 8.2 million binary reactions, which represents the bulk of all chemical reactions ever published in the scientific literature. Our model outperforms a rule‐based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data‐driven model generalises even beyond known reaction types, and is thus capable of effectively (re‐)discovering novel transformations (even including transition metal‐catalysed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub‐second time frame, the model can be used as a high‐throughput generator of reaction hypotheses for reaction discovery.
Learn from past experience: The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations (see figure). A model has been developed to mimic chemical reasoning by formalising reaction prediction as finding missing links in a knowledge graph.</description><subject>artificial intelligence</subject><subject>augmented scientific discovery</subject><subject>Chemical reactions</subject><subject>Chemistry</subject><subject>Chemists</subject><subject>computational chemistry</subject><subject>Computers</subject><subject>Expert systems</subject><subject>graph theory</subject><subject>Hypotheses</subject><subject>Knowledge representation</subject><subject>organic chemistry</subject><subject>Predictions</subject><subject>Reasoning</subject><subject>Transition metals</subject><issn>0947-6539</issn><issn>1521-3765</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><recordid>eNqFkE1Lw0AQhhdRbK1ePUrAi5fU2e9d8CKhWqFFET2H7e5GU9KkZhOl_96E1gpePA3MPPPw8iJ0jmGMAci1fferMQEsgHEuDtAQc4JjKgU_REPQTMaCUz1AJyEsAUALSo_RgEglCJNyiG7mlfNFkZdvUdKpcmuK6NmbUJX9qqmip9q73DaRKV30UH76sunvtsmrMpyio8wUwZ_t5gi93k1ekmk8e7x_SG5nsWUYRJwRcFho7KlbWCYdpwvjwGlCMmaUlYpjRjKqBLdglNBUS0UdGM-VoJkmdISutt51XX20PjTpKg-2i21KX7UhxYphqQXT0KGXf9Bl1dZll66jNAfFCe2p8ZaydRVC7bN0XecrU29SDGnfa9r3mu577R4udtp2sfJuj_8U2QF6C3zlhd_8o0uT6WT-K_8G_pKB5A</recordid><startdate>20170502</startdate><enddate>20170502</enddate><creator>Segler, Marwin H. S.</creator><creator>Waller, Mark P.</creator><general>Wiley Subscription Services, Inc</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>K9.</scope><scope>7X8</scope></search><sort><creationdate>20170502</creationdate><title>Modelling Chemical Reasoning to Predict and Invent Reactions</title><author>Segler, Marwin H. S. ; Waller, Mark P.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c4106-f20d1691e3dbc47d53bad0d922f4a8c785142f3865c0a86939783d0ae5863f923</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>artificial intelligence</topic><topic>augmented scientific discovery</topic><topic>Chemical reactions</topic><topic>Chemistry</topic><topic>Chemists</topic><topic>computational chemistry</topic><topic>Computers</topic><topic>Expert systems</topic><topic>graph theory</topic><topic>Hypotheses</topic><topic>Knowledge representation</topic><topic>organic chemistry</topic><topic>Predictions</topic><topic>Reasoning</topic><topic>Transition metals</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Segler, Marwin H. S.</creatorcontrib><creatorcontrib>Waller, Mark P.</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>MEDLINE - Academic</collection><jtitle>Chemistry : a European journal</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Segler, Marwin H. S.</au><au>Waller, Mark P.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Modelling Chemical Reasoning to Predict and Invent Reactions</atitle><jtitle>Chemistry : a European journal</jtitle><addtitle>Chemistry</addtitle><date>2017-05-02</date><risdate>2017</risdate><volume>23</volume><issue>25</issue><spage>6118</spage><epage>6128</epage><pages>6118-6128</pages><issn>0947-6539</issn><eissn>1521-3765</eissn><abstract>The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph containing 14.4 million molecules and 8.2 million binary reactions, which represents the bulk of all chemical reactions ever published in the scientific literature. Our model outperforms a rule‐based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data‐driven model generalises even beyond known reaction types, and is thus capable of effectively (re‐)discovering novel transformations (even including transition metal‐catalysed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub‐second time frame, the model can be used as a high‐throughput generator of reaction hypotheses for reaction discovery.
Learn from past experience: The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations (see figure). A model has been developed to mimic chemical reasoning by formalising reaction prediction as finding missing links in a knowledge graph.</abstract><cop>Germany</cop><pub>Wiley Subscription Services, Inc</pub><pmid>27862477</pmid><doi>10.1002/chem.201604556</doi><tpages>11</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0947-6539 |
ispartof | Chemistry : a European journal, 2017-05, Vol.23 (25), p.6118-6128 |
issn | 0947-6539 1521-3765 |
language | eng |
recordid | cdi_proquest_miscellaneous_1841796490 |
source | Wiley Online Library - AutoHoldings Journals |
subjects | artificial intelligence augmented scientific discovery Chemical reactions Chemistry Chemists computational chemistry Computers Expert systems graph theory Hypotheses Knowledge representation organic chemistry Predictions Reasoning Transition metals |
title | Modelling Chemical Reasoning to Predict and Invent Reactions |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T12%3A59%3A22IST&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=Modelling%20Chemical%20Reasoning%20to%20Predict%20and%20Invent%20Reactions&rft.jtitle=Chemistry%20:%20a%20European%20journal&rft.au=Segler,%20Marwin%20H.%20S.&rft.date=2017-05-02&rft.volume=23&rft.issue=25&rft.spage=6118&rft.epage=6128&rft.pages=6118-6128&rft.issn=0947-6539&rft.eissn=1521-3765&rft_id=info:doi/10.1002/chem.201604556&rft_dat=%3Cproquest_cross%3E1841796490%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=1895085230&rft_id=info:pmid/27862477&rfr_iscdi=true |