Predicting unknown viral hosts with Dynamic Positive-Unlabeled learning

Predicting unknown viral hosts with Dynamic Positive-Unlabeled learning | 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 Biological Sciences - Article Predicting unknown viral hosts with Dynamic Positive-Unlabeled learning Gabriele Pignalberi, Andrea Tonelli, Stefano Giagu, Moreno Di Marco This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7187859/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Most emerging infectious diseases originate from animals (i.e. zoonoses), but our knowledge of host-pathogen links remains scant. AI models have been used to predict unknown zoonotic hosts, but face challenges from biased data and the absence of confirmed negative host-pathogen associations. Here, we introduce the Dynamic Positive-Unlabeled (DPU) learning framework, an extension of classical Positive-Unlabeled learning that enables Graph Neural Networks to predict missing links in incomplete networks. DPU learning integrates a propensity score model that estimates the likelihood of observing existing links with a classifier that predicts true link existence. This approach corrects predictions to account for sampling bias and recognizes that missing links may result from either a true absence of association or gaps in data collection. We applied DPU learning to predict associations between 5,330 wild mammalian species and 33 viral families worldwide, leveraging phylogeographic relationships between mammals, observed mammal-virus association patterns, mammalian life-history traits, and genetic features of the viruses. The approach demonstrated high validation performances, providing unbiased and accurate estimation of pathogen distribution across species. DPU learning emerges a valuable tool to support strategic, data-driven surveillance activities for proactive zoonotic risk mitigation. Biological sciences/Ecology/Ecological epidemiology Physical sciences/Mathematics and computing/Computer science Full Text Additional Declarations There is NO Competing Interest. Supplementary Files Guide.doc Guide.doc ExtendedDataFigure1.png Extended Data Figure 1 ExtendedDataTable1.png Extended Data Table 1 ExtendedDataFigure6.png Extended Data 6 ExtendedDataFigure8.png Extended Data Figure 8 ExtendedDataFigure2.png Extended Data Figure 2 ExtendedDataFigure7.png Extended Data 7 ExtendedDataFigure3.png Extended Data 3 ExtendedDataFigure4.png Extended Data 4 ExtendedDataFigure5.png Extended Data 5 ExtendedDataTable2.xlsx Extended Data Table 2 SupplementaryInformation.pdf Supplementary Information Cite Share Download PDF Status: Under Review 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-7187859","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":494506111,"identity":"3dd31004-ac46-4539-93ce-d6f3dcab45eb","order_by":0,"name":"Gabriele Pignalberi","email":"data:image/png;base64,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","orcid":"","institution":"Sapienza University or Rome","correspondingAuthor":true,"prefix":"","firstName":"Gabriele","middleName":"","lastName":"Pignalberi","suffix":""},{"id":494506112,"identity":"b923b5f8-3ba4-4719-9f4c-f3aff9adc3d3","order_by":1,"name":"Andrea Tonelli","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Tonelli","suffix":""},{"id":494506113,"identity":"c6fdd796-a302-47da-bead-f2b1d1fdcb26","order_by":2,"name":"Stefano Giagu","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Stefano","middleName":"","lastName":"Giagu","suffix":""},{"id":494506114,"identity":"dc0eb21f-62f1-497e-b2fc-81f45cb679fe","order_by":3,"name":"Moreno Di Marco","email":"","orcid":"https://orcid.org/0000-0002-8902-4193","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Moreno","middleName":"Di","lastName":"Marco","suffix":""}],"badges":[],"createdAt":"2025-07-22 14:06:55","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7187859/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7187859/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88324353,"identity":"37a0e541-c311-478b-b4e9-dfcaa84743b2","added_by":"auto","created_at":"2025-08-05 09:30:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":676337,"visible":true,"origin":"","legend":"Article File","description":"","filename":"MainText.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1_covered_95e5258d-a6b6-4e28-9060-d68704b96442.pdf"},{"id":88322960,"identity":"2af8a5a6-9fc6-4645-89d7-39004b7dfceb","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":29696,"visible":true,"origin":"","legend":"Guide.doc","description":"","filename":"Guide.doc","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/405af379803b2ee0866b723e.doc"},{"id":88323376,"identity":"598b6207-3197-47bd-a57e-9655bf1e6821","added_by":"auto","created_at":"2025-08-05 09:22:47","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":1005205,"visible":true,"origin":"","legend":"Extended Data Figure 1","description":"","filename":"ExtendedDataFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/1ea3ec1fe4b698fbd6aa5c6f.png"},{"id":88322961,"identity":"788fb23c-e1a2-475a-8f22-4ce43f4b4ca9","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":829914,"visible":true,"origin":"","legend":"Extended Data Table 1","description":"","filename":"ExtendedDataTable1.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/278a3ad325ccbf3ff5b6f3b5.png"},{"id":88322962,"identity":"7f9d4c3f-a8f8-4493-918b-2aff228bc688","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":1882303,"visible":true,"origin":"","legend":"Extended Data 6","description":"","filename":"ExtendedDataFigure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/804edda1b5163e0cd02691c8.png"},{"id":88322964,"identity":"b5f47b65-6e8c-4a61-9561-d8842dd3db18","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":721314,"visible":true,"origin":"","legend":"Extended Data Figure 8","description":"","filename":"ExtendedDataFigure8.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/d3d7bee62ba8ff2fa9b9915c.png"},{"id":88322965,"identity":"2e64e788-fd6a-40a1-ac14-cfd1daaa1402","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":2244336,"visible":true,"origin":"","legend":"Extended Data Figure 2","description":"","filename":"ExtendedDataFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/e4aa8c5c5b885c98103cfe31.png"},{"id":88322969,"identity":"947e4ee3-679c-4077-b83a-74186195faf1","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":3297959,"visible":true,"origin":"","legend":"Extended Data 7","description":"","filename":"ExtendedDataFigure7.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/f827f232e6f69712fdf4d39c.png"},{"id":88322967,"identity":"77703cab-31f9-45de-8aa4-d848edad296d","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":3635203,"visible":true,"origin":"","legend":"Extended Data 3","description":"","filename":"ExtendedDataFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/e18d118048ccd59db34c5e89.png"},{"id":88323377,"identity":"55f2d5fb-a28e-46fe-9846-152e7413d67e","added_by":"auto","created_at":"2025-08-05 09:22:47","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":3821862,"visible":true,"origin":"","legend":"Extended Data 4","description":"","filename":"ExtendedDataFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/25288ac586544f3c08730e23.png"},{"id":88322966,"identity":"0d75439b-a13a-47e5-aecd-9f4f851741fc","added_by":"auto","created_at":"2025-08-05 09:14:47","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":625184,"visible":true,"origin":"","legend":"Extended Data 5","description":"","filename":"ExtendedDataFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/5b8c1d7865128c1097b8f8e0.png"},{"id":88322970,"identity":"d443eaba-73ce-4f56-9f30-4a19983eccaf","added_by":"auto","created_at":"2025-08-05 09:14:48","extension":"xlsx","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":41140493,"visible":true,"origin":"","legend":"Extended Data Table 2","description":"","filename":"ExtendedDataTable2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/1eb53a5c64d7a865fd41e608.xlsx"},{"id":88322971,"identity":"851e7505-d795-41ff-9cf8-77aad9ef2701","added_by":"auto","created_at":"2025-08-05 09:14:48","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":77626999,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7187859/v1/b123d68bf39610a25faab319.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Predicting unknown viral hosts with Dynamic Positive-Unlabeled learning","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-7187859/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7187859/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Most emerging infectious diseases originate from animals (i.e. zoonoses), but our knowledge of host-pathogen links remains scant. AI models have been used to predict unknown zoonotic hosts, but face challenges from biased data and the absence of confirmed negative host-pathogen associations. Here, we introduce the Dynamic Positive-Unlabeled (DPU) learning framework, an extension of classical Positive-Unlabeled learning that enables Graph Neural Networks to predict missing links in incomplete networks. DPU learning integrates a propensity score model that estimates the likelihood of observing existing links with a classifier that predicts true link existence. This approach corrects predictions to account for sampling bias and recognizes that missing links may result from either a true absence of association or gaps in data collection. We applied DPU learning to predict associations between 5,330 wild mammalian species and 33 viral families worldwide, leveraging phylogeographic relationships between mammals, observed mammal-virus association patterns, mammalian life-history traits, and genetic features of the viruses. The approach demonstrated high validation performances, providing unbiased and accurate estimation of pathogen distribution across species. DPU learning emerges a valuable tool to support strategic, data-driven surveillance activities for proactive zoonotic risk mitigation.","manuscriptTitle":"Predicting unknown viral hosts with Dynamic Positive-Unlabeled learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-05 09:14:42","doi":"10.21203/rs.3.rs-7187859/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e9840102-476f-4137-bd00-99c5ea97d45d","owner":[],"postedDate":"August 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":52523230,"name":"Biological sciences/Ecology/Ecological epidemiology"},{"id":52523231,"name":"Physical sciences/Mathematics and computing/Computer science"}],"tags":[],"updatedAt":"2025-10-23T10:21:44+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-05 09:14:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7187859","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7187859","identity":"rs-7187859","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}