Unsupervised Ensemble Learning for Efficient Integration of Pre-trained Polygenic Risk Scores

Unsupervised Ensemble Learning for Efficient Integration of Pre-trained Polygenic Risk Scores | 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 Unsupervised Ensemble Learning for Efficient Integration of Pre-trained Polygenic Risk Scores Rui Duan, Chenyin Gao, Justin Tubbs, Yi Han, Min Guo, Sijia Li, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5976048/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The growing availability of pre-trained polygenic risk score (PRS) models has enabled their integration into real-world applications, reducing the need for extensive data labeling, training, and calibration. However, selecting the most suitable PRS model for a specific target population remains challenging, due to issues such as limited transferability, data heterogeneity, and the scarcity of observed phenotype in real-world settings. Ensemble learning offers a promising avenue to enhance the predictive accuracy of genetic risk assessments, but most existing methods often rely on observed phenotype data or additional genome-wide association studies (GWAS) from the target population to optimize ensemble weights, limiting their utility in real-time implementation. Here, we present the UNSupervised enSemble PRS (UNSemblePRS), an unsupervised ensemble learning framework, that combines pre-trained PRS models without requiring phenotype data or summaries from the target population. Unlike traditional supervised approaches, UNSemblePRS aggregates models based on prediction concordance across a curated subset of candidate PRS models. We evaluated UNSemblePRS using both continuous and binary traits in the All of Us database, demonstrating its scalability and robust performance across diverse populations. These results underscore UNSemblePRS as an accessible tool for integrating PRS models into real-world contexts, offering broad applicability as the availability of PRS models continues to expand. Health sciences/Medical research/Genetics research Biological sciences/Genetics/Genetic association study/Genome-wide association studies Full Text Additional Declarations There is NO Competing Interest. Supplementary Files NGSupplementaryTable.pdf Metadata for PRS models from PGS catalog Cite Share Download PDF Status: Posted 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-5976048","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":435464327,"identity":"1a7c7543-c19b-455b-8220-270ce0118405","order_by":0,"name":"Rui Duan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYFCCBCCuOMDAIMHAhhDhIajlDMlaGNtI0WLenmP4uXDeHXtz6Qa2Bz/3HJY3b09gfPC2DbcWmTNvjKVnbnuWuHPOAXbDnmeHDeececBsOBePFgmJHANp3m2HEwxuJLBJ8BxIY5whkcAmzYtfi/Fv3jmH7UFaJP8cSLMHamH/TUCLmTRvw2HGDUAt0jwHbBJBtjDj1cLzrMya59jhxJ0zEtukZQ7YJM/gedgsOeccHi3syZtv89QctjeXSD4m+eaAhO0M9uSDH96U4dYCBwYMjA1QJpxBUMsoGAWjYBSMAhwAABZUUGJH7KCkAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-9261-4864","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":true,"prefix":"","firstName":"Rui","middleName":"","lastName":"Duan","suffix":""},{"id":435464328,"identity":"de4667ad-d7e6-42fc-9f48-b2ba722b830f","order_by":1,"name":"Chenyin Gao","email":"","orcid":"","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Chenyin","middleName":"","lastName":"Gao","suffix":""},{"id":435464329,"identity":"bb94f3ba-ca38-481c-abb1-0580a616e282","order_by":2,"name":"Justin Tubbs","email":"","orcid":"https://orcid.org/0000-0002-6671-6816","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Justin","middleName":"","lastName":"Tubbs","suffix":""},{"id":435464330,"identity":"c6e84a01-cabb-4da2-a756-a8ea845aae83","order_by":3,"name":"Yi Han","email":"","orcid":"","institution":"Department of Statistics, Columbia University","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Han","suffix":""},{"id":435464331,"identity":"56b08fdc-3f03-4a22-b65a-34cc80a0fcfb","order_by":4,"name":"Min Guo","email":"","orcid":"https://orcid.org/0009-0007-9652-3496","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"","lastName":"Guo","suffix":""},{"id":435464332,"identity":"f2342623-0db6-4bab-92ee-75ca18ea173d","order_by":5,"name":"Sijia Li","email":"","orcid":"","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Sijia","middleName":"","lastName":"Li","suffix":""},{"id":435464333,"identity":"eceaa8fb-0845-4cf3-be5c-f57ddc87e170","order_by":6,"name":"Erica Ma","email":"","orcid":"","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Erica","middleName":"","lastName":"Ma","suffix":""},{"id":435464334,"identity":"bc93c1db-43d4-4a07-9447-72c4020ea802","order_by":7,"name":"Dailin Luo","email":"","orcid":"","institution":"Department of Biostatistics, Harvard T.H. Chan School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Dailin","middleName":"","lastName":"Luo","suffix":""},{"id":435464335,"identity":"21500cf4-16dc-4387-986f-f9c6a7e64dec","order_by":8,"name":"Jordan Smoller","email":"","orcid":"","institution":"Massachusetts General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jordan","middleName":"","lastName":"Smoller","suffix":""},{"id":435464336,"identity":"b6e4ff65-b546-4e27-ad66-4f3bd36945cb","order_by":9,"name":"Phil Lee","email":"","orcid":"https://orcid.org/0000-0003-1770-3100","institution":"Massachusetts General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Phil","middleName":"","lastName":"Lee","suffix":""}],"badges":[],"createdAt":"2025-02-06 19:45:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5976048/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5976048/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85493770,"identity":"7b1fe2cf-4e60-480f-a076-96267fc06d31","added_by":"auto","created_at":"2025-06-26 13:24:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":997003,"visible":true,"origin":"","legend":"Article File","description":"","filename":"mainsupplement.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5976048/v1_covered_0062500f-8ff9-4f39-b49c-8dbe19a2cb9d.pdf"},{"id":79641780,"identity":"e64a509f-3a96-40bf-9115-5ae78a2c2687","added_by":"auto","created_at":"2025-04-01 06:12:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":133787,"visible":true,"origin":"","legend":"Metadata for PRS models from PGS catalog","description":"","filename":"NGSupplementaryTable.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5976048/v1/0e97452a4d60fe215e44e4c8.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Unsupervised Ensemble Learning for Efficient Integration of Pre-trained Polygenic Risk Scores","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5976048/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5976048/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The growing availability of pre-trained polygenic risk score (PRS) models has enabled their integration into real-world applications, reducing the need for extensive data labeling, training, and calibration. However, selecting the most suitable PRS model for a specific target population remains challenging, due to issues such as limited transferability, data heterogeneity, and the scarcity of observed phenotype in real-world settings. Ensemble learning offers a promising avenue to enhance the predictive accuracy of genetic risk assessments, but most existing methods often rely on observed phenotype data or additional genome-wide association studies (GWAS) from the target population to optimize ensemble weights, limiting their utility in real-time implementation. Here, we present the UNSupervised enSemble PRS (UNSemblePRS), an unsupervised ensemble learning framework, that combines pre-trained PRS models without requiring phenotype data or summaries from the target population. Unlike traditional supervised approaches, UNSemblePRS aggregates models based on prediction concordance across a curated subset of candidate PRS models. We evaluated UNSemblePRS using both continuous and binary traits in the All of Us database, demonstrating its scalability and robust performance across diverse populations. These results underscore UNSemblePRS as an accessible tool for integrating PRS models into real-world contexts, offering broad applicability as the availability of PRS models continues to expand.","manuscriptTitle":"Unsupervised Ensemble Learning for Efficient Integration of Pre-trained Polygenic Risk Scores","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-01 06:12:33","doi":"10.21203/rs.3.rs-5976048/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b726e07e-c598-4997-955b-638dd29a429c","owner":[],"postedDate":"April 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":46372056,"name":"Health sciences/Medical research/Genetics research"},{"id":46372057,"name":"Biological sciences/Genetics/Genetic association study/Genome-wide association studies"}],"tags":[],"updatedAt":"2025-06-26T13:15:58+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-01 06:12:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5976048","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5976048","identity":"rs-5976048","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}