Learning heritable multimodal brain representation via contrastive learning

Learning heritable multimodal brain representation via contrastive 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 Article Learning heritable multimodal brain representation via contrastive learning Degui Zhi, Tian Xia, Xingzhong Zhao, Sheikh Muhammad Saiful Islam, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8866615/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 Magnetic resonance imaging (MRI)-derived phenotypes (IDP) has enabled the discovery of numerous genomic loci associated with brain structure and function. However, most existing IDPs and learned representations are derived from a single imaging modality, missing complementary information across modalities and potentially limiting the scope of genetic discovery. Here, we introduce a multimodal contrastive learning framework to derive heritable representations from paired T1- and T2-weighted MRIs. Unlike single-modality reconstruction-based models, we designed a momentum-based contrastive learning framework. The derived heritable representations show improved correlation with traditional IDPs and better predict age and brain disorders. Notably, genome-wide association studies (GWAS) of the learned representations reveal a substantially higher overlap of genetic loci across modalities, indicating improved alignment of their underlying genetic architecture. Analysis of the GWAS loci identified shared protein and drug targets, yielding meaningful biological insights. Overall, our framework learns shared representations across brain imaging modalities that exhibit anatomical and genetic coherence. Biological sciences/Computational biology and bioinformatics/Computational models Biological sciences/Genetics/Heritable quantitative trait/Quantitative trait loci Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Computational biology and bioinformatics/Image processing Full Text Additional Declarations There is NO Competing Interest. Supplementary Files mocov2SuppTableCompiled.xlsx Supplementary Tables mocov2SupplementalMaterials.docx Supplementary Notes and Figures 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-8866615","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":628491336,"identity":"ddd62a87-04bd-479b-a2cf-8e9ddbf34f88","order_by":0,"name":"Degui Zhi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYBACexCRwPAvgY29gYEZJiqBT4thA1gLWwIbzwHGZgYGA8JaDA6AKbYEBokEYrUcb78m8YCBJ49P8o354wKGP9H8DcwHb/Pg03LmTJkE0IpiNukcw+YZDAa5Mw6wJVvj1XIjJw2oxSCxDaSFB6hlAwOPmTReLfffgLQkJLZJnoFp4f+GX8sN9mNALQcS2yR44Law4dVi2JPDbJFgANTCk1Y4m8fAOHfGYTZjyzl4tNizH39480fFv8T57Yc3fOapkMvtb29+eOMNHi0MDDwGsNhggDCY8SiGAPYHBJWMglEwCkbBCAcAYMtHXZ6DwfAAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-7754-1890","institution":"The University of Texas Health Science Center at Houston","correspondingAuthor":true,"prefix":"","firstName":"Degui","middleName":"","lastName":"Zhi","suffix":""},{"id":628491337,"identity":"949f363c-1c69-4296-9617-ad566afa082b","order_by":1,"name":"Tian Xia","email":"","orcid":"","institution":"The University of Texas Health Science Center at Houston","correspondingAuthor":false,"prefix":"","firstName":"Tian","middleName":"","lastName":"Xia","suffix":""},{"id":628491338,"identity":"262eb48c-3015-4447-809b-aaee3554e8b6","order_by":2,"name":"Xingzhong Zhao","email":"","orcid":"https://orcid.org/0000-0002-7508-0856","institution":"The University of Texas Health Science Center at Houston","correspondingAuthor":false,"prefix":"","firstName":"Xingzhong","middleName":"","lastName":"Zhao","suffix":""},{"id":628491339,"identity":"bb69bcb5-9167-4778-96e2-1ab6fe87a4ae","order_by":3,"name":"Sheikh Muhammad Saiful Islam","email":"","orcid":"","institution":"The University of Texas Health Science Center at Houston","correspondingAuthor":false,"prefix":"","firstName":"Sheikh","middleName":"Muhammad Saiful","lastName":"Islam","suffix":""},{"id":628491340,"identity":"7bf484ef-126a-4189-9ad7-b5d71065a6b0","order_by":4,"name":"Kamil Mohammed","email":"","orcid":"","institution":"uthealth houston","correspondingAuthor":false,"prefix":"","firstName":"Kamil","middleName":"","lastName":"Mohammed","suffix":""},{"id":628491341,"identity":"3c07cda7-4267-4b53-8f85-2d31cf77ebf5","order_by":5,"name":"Ziqian Xie","email":"","orcid":"https://orcid.org/0000-0001-6541-1773","institution":"The University of Texas Health Science Center at Houston","correspondingAuthor":false,"prefix":"","firstName":"Ziqian","middleName":"","lastName":"Xie","suffix":""}],"badges":[],"createdAt":"2026-02-13 02:30:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8866615/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8866615/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108182749,"identity":"afe05e44-8b8b-458e-abd1-111ef412206e","added_by":"auto","created_at":"2026-04-30 08:59:32","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1535285,"visible":true,"origin":"","legend":"","description":"","filename":"mocov2paperTianrevisionmarkup.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8866615/v1_covered_50865dc1-cb99-42f4-a263-0451b7658204.pdf"},{"id":107694438,"identity":"f82641c2-1da6-4953-8be3-bbeedce7ee41","added_by":"auto","created_at":"2026-04-24 06:42:49","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":101682,"visible":true,"origin":"","legend":"Supplementary Tables","description":"","filename":"mocov2SuppTableCompiled.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8866615/v1/8d83aa1a32b1fa6c41f2ac89.xlsx"},{"id":107694439,"identity":"620a4405-e34a-47e0-81bd-6f9811ddad35","added_by":"auto","created_at":"2026-04-24 06:42:49","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":5445778,"visible":true,"origin":"","legend":"Supplementary Notes and Figures","description":"","filename":"mocov2SupplementalMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-8866615/v1/d2e5919738136a9b37c646d2.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Learning heritable multimodal brain representation via contrastive learning","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-8866615/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8866615/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Magnetic resonance imaging (MRI)-derived phenotypes (IDP) has enabled the discovery of numerous genomic loci associated with brain structure and function. However, most existing IDPs and learned representations are derived from a single imaging modality, missing complementary information across modalities and potentially limiting the scope of genetic discovery. Here, we introduce a multimodal contrastive learning framework to derive heritable representations from paired T1- and T2-weighted MRIs. Unlike single-modality reconstruction-based models, we designed a momentum-based contrastive learning framework. The derived heritable representations show improved correlation with traditional IDPs and better predict age and brain disorders. Notably, genome-wide association studies (GWAS) of the learned representations reveal a substantially higher overlap of genetic loci across modalities, indicating improved alignment of their underlying genetic architecture. Analysis of the GWAS loci identified shared protein and drug targets, yielding meaningful biological insights. Overall, our framework learns shared representations across brain imaging modalities that exhibit anatomical and genetic coherence.","manuscriptTitle":"Learning heritable multimodal brain representation via contrastive learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-24 06:42:45","doi":"10.21203/rs.3.rs-8866615/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":"e87993a6-8055-4783-9341-62c36713b0c9","owner":[],"postedDate":"April 24th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Reject before peer review","date":"2026-04-29T22:04:24+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66896080,"name":"Biological sciences/Computational biology and bioinformatics/Computational models"},{"id":66896081,"name":"Biological sciences/Genetics/Heritable quantitative trait/Quantitative trait loci"},{"id":66896082,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":66896083,"name":"Biological sciences/Computational biology and bioinformatics/Image processing"}],"tags":[],"updatedAt":"2026-04-29T22:05:20+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-24 06:42:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8866615","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8866615","identity":"rs-8866615","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}