Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach

Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach | 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 Research Article Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach Adel Memariani, Martin Glauer, Simon Flügel, Fabian Neuhaus, Janna Hastings, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6546788/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 01 Sep, 2025 Read the published version in Journal of Cheminformatics → Version 1 posted 9 You are reading this latest preprint version Abstract Deriving symbolic knowledge from trained deep learning models is challenging due to the lack of transparency in models. A promising approach to address this issue is to couple a semantic structure with the predictions and make the outcomes interpretable. In prediction tasks such as multi-label classification, labels tend to form hierarchical relationships. Therefore, we propose enforcing a taxonomical structure on the model’s outputs throughout the training phase. In vector space, a taxonomy can be represented using axis-aligned hyperrectangles, or boxes, which may overlap or nest within one another. The boundaries of a box determine the extent of a particular category. Thus, we used box-shaped embeddings of ontology classes to learn and transparently represent logical relations that are only implicit in multi-label datasets. We assessed our model by measuring its ability to approximate the deductive closure of subsumption relations in the ChEBI ontology, which is a distinguished knowledge base in the field of chemistry. We demonstrate that our model captures implicit hierarchical relationships among labels, ensuring consistency with the underlying ontological conceptualization, while also achieving state-of-the-art performance in multi-label classification. Notably, this is accomplished without requiring an explicit taxonomy during the training process. Scientific Contribution: Our proposed approach advances chemical classification by enabling interpretable outputs through a structured and geometrically expressive representation of molecules and their classes. Ontology Box Embedding Classification ChEBI Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 01 Sep, 2025 Read the published version in Journal of Cheminformatics → Version 1 posted Editorial decision: Revision requested 30 Jul, 2025 Reviews received at journal 27 Jul, 2025 Reviews received at journal 11 Jun, 2025 Reviewers agreed at journal 28 May, 2025 Reviewers agreed at journal 20 May, 2025 Reviewers invited by journal 08 May, 2025 Editor assigned by journal 06 May, 2025 Submission checks completed at journal 06 May, 2025 First submitted to journal 28 Apr, 2025 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-6546788","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":454732564,"identity":"73959244-92f9-48ad-a47d-fef82735e904","order_by":0,"name":"Adel Memariani","email":"data:image/png;base64,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","orcid":"","institution":"Paderborn University","correspondingAuthor":true,"prefix":"","firstName":"Adel","middleName":"","lastName":"Memariani","suffix":""},{"id":454732565,"identity":"f4b6cb4b-a96d-4564-9330-3f38521095e3","order_by":1,"name":"Martin Glauer","email":"","orcid":"","institution":"Otto von Guericke University","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Glauer","suffix":""},{"id":454732566,"identity":"f4f8401c-a192-48a0-91fc-5ecefa336857","order_by":2,"name":"Simon Flügel","email":"","orcid":"","institution":"Osnabrück University","correspondingAuthor":false,"prefix":"","firstName":"Simon","middleName":"","lastName":"Flügel","suffix":""},{"id":454732567,"identity":"4f14dbbd-0aef-41c4-8baf-e4a1fa366ce9","order_by":3,"name":"Fabian Neuhaus","email":"","orcid":"","institution":"Otto von Guericke University","correspondingAuthor":false,"prefix":"","firstName":"Fabian","middleName":"","lastName":"Neuhaus","suffix":""},{"id":454732570,"identity":"384858ba-f4ec-410d-97ce-893a15e6998e","order_by":4,"name":"Janna Hastings","email":"","orcid":"","institution":"Faculty of Medicine, University of Zurich","correspondingAuthor":false,"prefix":"","firstName":"Janna","middleName":"","lastName":"Hastings","suffix":""},{"id":454732572,"identity":"2e34b780-545f-4cba-a2f2-d0d0581b2905","order_by":5,"name":"Till Mossakowski","email":"","orcid":"","institution":"Osnabrück University","correspondingAuthor":false,"prefix":"","firstName":"Till","middleName":"","lastName":"Mossakowski","suffix":""}],"badges":[],"createdAt":"2025-04-28 10:23:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6546788/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6546788/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13321-025-01086-1","type":"published","date":"2025-09-01T15:57:15+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":90827933,"identity":"7b949ecd-f5ec-4923-974e-02f1d5615b19","added_by":"auto","created_at":"2025-09-08 16:03:38","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2586298,"visible":true,"origin":"","legend":"","description":"","filename":"BoxEmbeddingsforExtendingOntologies.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6546788/v1_covered_2509bef6-f257-440a-b510-0de06be5eaba.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"journal-of-cheminformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"chin","sideBox":"Learn more about [Journal of Cheminformatics](https://jcheminf.biomedcentral.com/)","snPcode":"13321","submissionUrl":"https://submission.nature.com/new-submission/13321/3","title":"Journal of Cheminformatics","twitterHandle":"@jcheminf","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Ontology, Box Embedding, Classification, ChEBI","lastPublishedDoi":"10.21203/rs.3.rs-6546788/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6546788/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDeriving symbolic knowledge from trained deep learning models is challenging due to the lack of transparency in models. A promising approach to address this issue is to couple a semantic structure with the predictions and make the outcomes interpretable. In prediction tasks such as multi-label classification, labels tend to form hierarchical relationships. Therefore, we propose enforcing a taxonomical structure on the model’s outputs throughout the training phase. In vector space, a taxonomy can be represented using axis-aligned hyperrectangles, or boxes, which may overlap or nest within one another. The boundaries of a box determine the extent of a particular category. Thus, we used box-shaped embeddings of ontology classes to learn and transparently represent logical relations that are only implicit in multi-label datasets. We assessed our model by measuring its ability to approximate the deductive closure of subsumption relations in the ChEBI ontology, which is a distinguished knowledge base in the field of chemistry. We demonstrate that our model captures implicit hierarchical relationships among labels, ensuring consistency with the underlying ontological conceptualization, while also achieving state-of-the-art performance in multi-label classification. Notably, this is accomplished without requiring an explicit taxonomy during the training process.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eScientific Contribution:\u003c/strong\u003e Our proposed approach advances chemical classification by enabling interpretable outputs through a structured and geometrically expressive representation of molecules and their classes.\u003c/p\u003e","manuscriptTitle":"Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-13 10:25:10","doi":"10.21203/rs.3.rs-6546788/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-30T05:24:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-27T15:11:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-11T10:57:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"259248169406331471745625592671733670121","date":"2025-05-28T08:40:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"310763252518712311271173761499227785723","date":"2025-05-20T10:45:54+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-08T14:08:38+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-06T08:07:33+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-06T08:04:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Cheminformatics","date":"2025-04-28T10:17:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-cheminformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"chin","sideBox":"Learn more about [Journal of Cheminformatics](https://jcheminf.biomedcentral.com/)","snPcode":"13321","submissionUrl":"https://submission.nature.com/new-submission/13321/3","title":"Journal of Cheminformatics","twitterHandle":"@jcheminf","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c8d778b6-4f62-4966-b95f-b100fa92a64a","owner":[],"postedDate":"May 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-09-08T15:59:30+00:00","versionOfRecord":{"articleIdentity":"rs-6546788","link":"https://doi.org/10.1186/s13321-025-01086-1","journal":{"identity":"journal-of-cheminformatics","isVorOnly":false,"title":"Journal of Cheminformatics"},"publishedOn":"2025-09-01 15:57:15","publishedOnDateReadable":"September 1st, 2025"},"versionCreatedAt":"2025-05-13 10:25:10","video":"","vorDoi":"10.1186/s13321-025-01086-1","vorDoiUrl":"https://doi.org/10.1186/s13321-025-01086-1","workflowStages":[]},"version":"v1","identity":"rs-6546788","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6546788","identity":"rs-6546788","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}