DeepGO-SE: Protein function prediction as Approximate Semantic Entailment

DeepGO-SE: Protein function prediction as Approximate Semantic Entailment | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article DeepGO-SE: Protein function prediction as Approximate Semantic Entailment Maxat Kulmanov, Francisco Guzmán-Vega, Paula Duek, Lydie Lane, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3258432/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Feb, 2024 Read the published version in Nature Machine Intelligence → Version 1 posted You are reading this latest preprint version Abstract The Gene Ontology (GO) is one of the most successful ontologies in the biological domain. GO is a formal theory with over 100,000 axioms that describe the molecular functions, biological processes, and cellular locations of proteins in three sub-ontologies. Many methods have been developed to automatically predict protein functions. However, only few of them use the background knowledge provided in the axioms of GO for knowledge-enhanced machine learning, or adjust and evaluate the model for the differences between the sub-ontologies. We have developed DeepGO-SE, a novel method which predicts GO functions from protein sequences using a pretrained large language model combined with a neuro-symbolic model that exploits GO axioms and performs protein function prediction as a form of approximate semantic entailment. We specifically evaluate DeepGO-SE on proteins that have no significant similarity with training proteins and demonstrate that DeepGO-SE can improve function prediction for those proteins. Biological sciences/Computational biology and bioinformatics/Protein function predictions Biological sciences/Computational biology and bioinformatics/Gene ontology Biological sciences/Computational biology and bioinformatics/Machine learning Protein Function Gene Ontology Ontology Embedding Machine Learning Full Text Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Published Journal Publication published 14 Feb, 2024 Read the published version in Nature Machine Intelligence → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3258432","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":236169382,"identity":"079cd9e7-de7b-4a8b-b1a4-0db12793a81b","order_by":0,"name":"Maxat Kulmanov","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-1710-1820","institution":"King Abdullah University of Science and Technology","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Maxat","middleName":"","lastName":"Kulmanov","suffix":""},{"id":236169383,"identity":"f9e3b9bc-68fc-48b4-808e-3e0882699f92","order_by":1,"name":"Francisco Guzmán-Vega","email":"","orcid":"","institution":"King Abdullah University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Francisco","middleName":"","lastName":"Guzmán-Vega","suffix":""},{"id":236169384,"identity":"27e01f83-a37b-4b5b-b141-ab31dcad78a1","order_by":2,"name":"Paula Duek","email":"","orcid":"","institution":"SIB-Swiss Institute of Bioinformatics","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Paula","middleName":"","lastName":"Duek","suffix":""},{"id":236169385,"identity":"a69cf2f1-110f-4248-9028-1a26371df0ae","order_by":3,"name":"Lydie Lane","email":"","orcid":"","institution":"SIB-Swiss Institute of Bioinformatics","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lydie","middleName":"","lastName":"Lane","suffix":""},{"id":236169386,"identity":"62ce9f14-64a3-4905-b2b0-b1cd0ce191a6","order_by":4,"name":"Stefan Arold","email":"","orcid":"","institution":"King Abdullah University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Stefan","middleName":"","lastName":"Arold","suffix":""},{"id":236169387,"identity":"a9b3f747-e37d-411a-bf11-7edc9f69275c","order_by":5,"name":"Robert Hoehndorf","email":"","orcid":"","institution":"King Abdullah University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Robert","middleName":"","lastName":"Hoehndorf","suffix":""}],"badges":[],"createdAt":"2023-08-12 14:55:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3258432/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3258432/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s42256-024-00795-w","type":"published","date":"2024-02-14T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":51162567,"identity":"06675c78-3443-4db4-9345-1fd9b0a03cac","added_by":"auto","created_at":"2024-02-15 08:08:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":995903,"visible":true,"origin":"","legend":"","description":"","filename":"DeepGOSE.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3258432/v1_covered_ac3784a7-f603-44dc-9182-0b9993eed765.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"DeepGO-SE: Protein function prediction as\r\nApproximate Semantic Entailment","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Protein Function, Gene Ontology, Ontology Embedding, Machine Learning","lastPublishedDoi":"10.21203/rs.3.rs-3258432/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3258432/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Gene Ontology (GO) is one of the most successful ontologies in the biological domain. GO is a formal theory with over 100,000 axioms that describe the molecular functions, biological processes, and cellular locations of proteins in three sub-ontologies. Many methods have been developed to automatically predict protein functions. However, only few of them use the background knowledge provided in the axioms of GO for knowledge-enhanced machine learning, or adjust and evaluate the model for the differences between the sub-ontologies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe have developed DeepGO-SE, a novel method which predicts GO functions from protein sequences using a pretrained large language model combined with a neuro-symbolic model that exploits GO axioms and performs protein function prediction as a form of approximate semantic entailment. We specifically evaluate DeepGO-SE on proteins that have no significant similarity with training proteins and demonstrate that DeepGO-SE can improve function prediction for those proteins.\u003c/p\u003e","manuscriptTitle":"DeepGO-SE: Protein function prediction as\nApproximate Semantic Entailment","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-09-27 02:33:51","doi":"10.21203/rs.3.rs-3258432/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-machine-intelligence","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natmachintell","sideBox":"Learn more about [Nature Machine Intelligence](http://www.nature.com/natmachintell/)","snPcode":"","submissionUrl":"","title":"Nature Machine Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"ef4dc138-ddca-4be6-b88f-b6d7c3d738bf","owner":[],"postedDate":"September 27th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":25009961,"name":"Biological sciences/Computational biology and bioinformatics/Protein function predictions"},{"id":25009962,"name":"Biological sciences/Computational biology and bioinformatics/Gene ontology"},{"id":25009963,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"}],"tags":[],"updatedAt":"2024-02-15T08:07:52+00:00","versionOfRecord":{"articleIdentity":"rs-3258432","link":"https://doi.org/10.1038/s42256-024-00795-w","journal":{"identity":"nature-machine-intelligence","isVorOnly":false,"title":"Nature Machine Intelligence"},"publishedOn":"2024-02-14 05:00:00","publishedOnDateReadable":"February 14th, 2024"},"versionCreatedAt":"2023-09-27 02:33:51","video":"","vorDoi":"10.1038/s42256-024-00795-w","vorDoiUrl":"https://doi.org/10.1038/s42256-024-00795-w","workflowStages":[]},"version":"v1","identity":"rs-3258432","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3258432","identity":"rs-3258432","version":["v1"]},"buildId":"cBFmMYwuxLRRLfASyISRj","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}