Neuronal and microglial fine-tuned 3D DNA foundation models dissect genetic variants in Alzheimer’s disease

Neuronal and microglial fine-tuned 3D DNA foundation models dissect genetic variants in Alzheimer’s disease | 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 Neuronal and microglial fine-tuned 3D DNA foundation models dissect genetic variants in Alzheimer’s disease Feixiong Cheng, Yunxiao Ren, Maria Paz, Xiaoyu Yang, Yichen Li, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8920766/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 Although genetic variants are among the strongest risk factors for Alzheimer’s disease (AD), most associated variants reside in non-coding regions with unclear mechanisms. Here, we address that gap by fine-tuning the Akita DNA foundation model on high-resolution Hi-C data generated from neuronal progenitor cells (NPCs) and microglia to produce iNeuroAkita , a cell-context-aware three-dimensional (3D) genome model. iNeuroAkita shows improved correlation with experimental Hi-C contact maps and more accurate contact predictions at key AD loci (e.g. APOE and PSEN2) , and captures CTCF-dependent chromatin loops. Using in silico saturation mutagenesis (ISM) across 11,856 AD-associated single-nucleotide variants (SNVs) and 120 structural variants (SVs), we quantified impacts on chromatin spatial organization and identified two SNVs predicted to strongly disrupt 3D chromatin architecture: rs636317 at the MS4A locus that is predicted to disrupt local chromatic interactions and microglia eQTL rs7599054 that creates gained interactions connecting its surrounding region to CCNT2 and TMEM163 , consistent with enhancer–promoter activation in epigenomic and single-cell transcriptomic data. We also identified AD-associated deletions at the KANSL1 and ARMS2 that rewire local regulatory contacts and associate with cell-type-specific expression changes, demonstrating iNeuroAkita’s utility for systematic characterization of noncoding AD risk variants underlying 3D genome regulation. Biological sciences/Genetics/Gene regulation Biological sciences/Computational biology and bioinformatics Health sciences/Neurology/Neurological disorders/Dementia/Alzheimer's disease Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTableS1.zip Supplementary Table 1 SupplementaryTableS2.zip Supplementary Table 2 SupplementaryFigS1.pdf Supplementary Figure 1 SupplementaryTableS8.zip Supplementary Table 8 SupplementaryTableS9.zip Supplementary Table 9 SupplementaryTableS3.zip Supplementary Table 3 SupplementaryTableS4.zip Supplementary Table 4 SupplementaryTableS7.zip Supplementary Table 7 SupplementaryFigS3.pdf Supplementary Figure 3 SupplementaryTableS5.zip Supplementary Table 5 SupplementaryFigS2.pdf Supplementary Figure 2 SupplementaryTableS6.zip Supplementary Table 6 SupplementaryFigS5.pdf Supplementary Figure 5 SupplementaryFigS4.pdf Supplementary Figure 4 SupplementaryTableLegends.pdf Supplementary Table Legends SupplementaryFigureLegends.pdf Supplementary Figure Legends 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-8920766","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":597064500,"identity":"06d36aec-f022-4532-8916-f1ce78839ac5","order_by":0,"name":"Feixiong Cheng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAt0lEQVRIiWNgGAWjYDACCRBhY8PAwHyAJC1paQwMbAmkaTlMghb+2c3HHn5JOC/P38bA+LjiFzGW3DmWbiyTcNtwxjEGZsOzfURoMZDIMZOW/HE7wUC+gU2ysYcoLfnfpCUSziUYsDEQrSWHTfJDwgGIloYfRGiRuJFmJs2QkAz0C2OzYWMDEVr4ZyQ/k/yRYAcMMeaDDxv+EKEFBJh5wBRjAwNjG5FaGBE+INaWUTAKRsEoGFEAAKT8MmqgJc+hAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-1736-2847","institution":"Cleveland Clinic","correspondingAuthor":true,"prefix":"","firstName":"Feixiong","middleName":"","lastName":"Cheng","suffix":""},{"id":597064501,"identity":"08dfd798-f20d-4769-a38e-6b487ec514f8","order_by":1,"name":"Yunxiao Ren","email":"","orcid":"","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Yunxiao","middleName":"","lastName":"Ren","suffix":""},{"id":597064502,"identity":"dbf6ee2d-b8a6-47de-a31a-34bd78887500","order_by":2,"name":"Maria Paz","email":"","orcid":"https://orcid.org/0000-0003-0862-5673","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Paz","suffix":""},{"id":597064503,"identity":"adf1b86f-0c4c-4543-a05b-5d1446687adf","order_by":3,"name":"Xiaoyu Yang","email":"","orcid":"","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyu","middleName":"","lastName":"Yang","suffix":""},{"id":597064504,"identity":"3c33db73-f021-47ea-a58a-d7d6e119e33d","order_by":4,"name":"Yichen Li","email":"","orcid":"https://orcid.org/0009-0005-1355-5018","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Yichen","middleName":"","lastName":"Li","suffix":""},{"id":597064505,"identity":"78ebb404-af03-4100-8a07-c3c378aa2d47","order_by":5,"name":"Yuan Hou","email":"","orcid":"","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Yuan","middleName":"","lastName":"Hou","suffix":""},{"id":597064506,"identity":"3922c62d-d17b-4e96-b834-400ebe1968e2","order_by":6,"name":"Yang Li","email":"","orcid":"https://orcid.org/0000-0001-6997-6018","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Li","suffix":""},{"id":597064507,"identity":"d9721207-3a31-4da2-9602-c8c57092ac31","order_by":7,"name":"Fulai Jin","email":"","orcid":"https://orcid.org/0000-0003-0025-4337","institution":"Case Western Reserve University","correspondingAuthor":false,"prefix":"","firstName":"Fulai","middleName":"","lastName":"Jin","suffix":""},{"id":597064508,"identity":"1c6dc917-1b26-46cd-a960-0751c216d892","order_by":8,"name":"Ming Hu","email":"","orcid":"https://orcid.org/0000-0003-0987-2916","institution":"Cleveland Clinic Lerner research institute","correspondingAuthor":false,"prefix":"","firstName":"Ming","middleName":"","lastName":"Hu","suffix":""},{"id":597064509,"identity":"a7290b15-5238-4112-b17e-8f95860b8525","order_by":9,"name":"Andrew Pieper","email":"","orcid":"","institution":"Case Western Reserve University","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Pieper","suffix":""}],"badges":[],"createdAt":"2026-02-19 20:50:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8920766/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8920766/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104407370,"identity":"d70014a4-a490-4739-b6ac-95b6862cdce6","added_by":"auto","created_at":"2026-03-11 12:37:23","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1356102,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptCheng.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1_covered_f88623a5-f774-4e83-a27b-6e2a47208d04.pdf"},{"id":103531683,"identity":"84fb99f9-8415-475c-b6b5-bc0dc6ddbb14","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":7755,"visible":true,"origin":"","legend":"Supplementary Table 1","description":"","filename":"SupplementaryTableS1.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/3dbadc7e9e530987c269030a.zip"},{"id":103531684,"identity":"97de9916-364a-493c-ba67-d436ba76b35a","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":156943,"visible":true,"origin":"","legend":"Supplementary Table 2","description":"","filename":"SupplementaryTableS2.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/1a6d97a921465471561fd9be.zip"},{"id":103531691,"identity":"2fc7434c-c0b8-4129-9d47-ba814c32c7be","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":843774,"visible":true,"origin":"","legend":"Supplementary Figure 1","description":"","filename":"SupplementaryFigS1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/15c4e783f73746221edbd0ab.pdf"},{"id":103531693,"identity":"8de905c0-bb31-456d-b60b-021f936af637","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":15663,"visible":true,"origin":"","legend":"Supplementary Table 8","description":"","filename":"SupplementaryTableS8.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/b6008da10edfe997b8eb5f77.zip"},{"id":103531688,"identity":"c21284de-1132-4979-9502-e0d0f57a3b0e","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":58214,"visible":true,"origin":"","legend":"Supplementary Table 9","description":"","filename":"SupplementaryTableS9.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/073deb50284da138221dd623.zip"},{"id":103531685,"identity":"6d9bfdb8-f1de-4480-9c23-1e4042a88431","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":29171,"visible":true,"origin":"","legend":"Supplementary Table 3","description":"","filename":"SupplementaryTableS3.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/0af7a533deb7b3e139f7ad13.zip"},{"id":104397680,"identity":"173e5a29-50b1-4cdd-ac96-0042f834001d","added_by":"auto","created_at":"2026-03-11 11:54:30","extension":"zip","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":11334,"visible":true,"origin":"","legend":"Supplementary Table 4","description":"","filename":"SupplementaryTableS4.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/6bcbf368ff4776f48264dead.zip"},{"id":104397837,"identity":"812c2040-32ad-44c7-9d81-a7ca295fb091","added_by":"auto","created_at":"2026-03-11 11:57:30","extension":"zip","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":17587,"visible":true,"origin":"","legend":"Supplementary Table 7","description":"","filename":"SupplementaryTableS7.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/e7ae44cb681810b55a6bf4e2.zip"},{"id":103531694,"identity":"77e0a9d0-868e-4127-81f6-7eb5a72d0404","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":8804,"visible":true,"origin":"","legend":"Supplementary Figure 3","description":"","filename":"SupplementaryFigS3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/78c528a1e9a53a2615192799.pdf"},{"id":103531695,"identity":"326b75f9-23d6-4506-b54a-854b858b4957","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"zip","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":16882,"visible":true,"origin":"","legend":"Supplementary Table 5","description":"","filename":"SupplementaryTableS5.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/088ac9369da9e6e4371603d8.zip"},{"id":104397751,"identity":"ec97de4b-0240-463e-b356-88519d305ffb","added_by":"auto","created_at":"2026-03-11 11:55:27","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":574205,"visible":true,"origin":"","legend":"Supplementary Figure 2","description":"","filename":"SupplementaryFigS2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/a877f16af475e26f2ec9beb7.pdf"},{"id":103531697,"identity":"af838486-07c0-4abd-ac4b-5d445e3f7148","added_by":"auto","created_at":"2026-02-26 17:12:35","extension":"zip","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":17576,"visible":true,"origin":"","legend":"Supplementary Table 6","description":"","filename":"SupplementaryTableS6.zip","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/007f76abd8cc6b75e1c506de.zip"},{"id":103531696,"identity":"1b36d4d5-69ab-40a4-8bf0-a74411ce6de9","added_by":"auto","created_at":"2026-02-26 17:12:34","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":130909,"visible":true,"origin":"","legend":"Supplementary Figure 5","description":"","filename":"SupplementaryFigS5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/352be66b70ce883bb54e8263.pdf"},{"id":103531700,"identity":"bc48dc7f-45e3-41da-9c23-76fdf3f5938f","added_by":"auto","created_at":"2026-02-26 17:12:35","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":8454,"visible":true,"origin":"","legend":"Supplementary Figure 4","description":"","filename":"SupplementaryFigS4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/0f85a9f0276409f7873828a9.pdf"},{"id":104397657,"identity":"365b93e9-6b2d-4de5-b526-fae35f0b4830","added_by":"auto","created_at":"2026-03-11 11:53:58","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":56553,"visible":true,"origin":"","legend":"Supplementary Table Legends","description":"","filename":"SupplementaryTableLegends.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/d04f529e658adcc1ac9723d1.pdf"},{"id":103531698,"identity":"2c8c3d60-5e42-4f36-818a-0afe492c26ac","added_by":"auto","created_at":"2026-02-26 17:12:35","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":26613,"visible":true,"origin":"","legend":"Supplementary Figure Legends","description":"","filename":"SupplementaryFigureLegends.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8920766/v1/a857b4af9a240b9b9a313422.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Neuronal and microglial fine-tuned 3D DNA foundation models dissect genetic variants in Alzheimer’s disease","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":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":"","lastPublishedDoi":"10.21203/rs.3.rs-8920766/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8920766/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAlthough genetic variants are among the strongest risk factors for Alzheimer\u0026rsquo;s disease (AD), most associated variants reside in non-coding regions with unclear mechanisms. Here, we address that gap by fine-tuning the Akita DNA foundation model on high-resolution Hi-C data generated from neuronal progenitor cells (NPCs) and microglia to produce \u003cem\u003eiNeuroAkita\u003c/em\u003e, a cell-context-aware three-dimensional (3D) genome model. \u003cem\u003eiNeuroAkita\u003c/em\u003e shows improved correlation with experimental Hi-C contact maps and more accurate contact predictions at key AD loci (e.g. \u003cem\u003eAPOE\u003c/em\u003e and \u003cem\u003ePSEN2)\u003c/em\u003e, and captures CTCF-dependent chromatin loops. Using \u003cem\u003ein silico\u003c/em\u003e saturation mutagenesis (ISM) across 11,856 AD-associated single-nucleotide variants (SNVs) and 120 structural variants (SVs), we quantified impacts on chromatin spatial organization and identified two SNVs predicted to strongly disrupt 3D chromatin architecture: rs636317 at the \u003cem\u003eMS4A\u003c/em\u003e locus that is predicted to disrupt local chromatic interactions and microglia eQTL rs7599054 that creates gained interactions connecting its surrounding region to \u003cem\u003eCCNT2\u003c/em\u003e and \u003cem\u003eTMEM163\u003c/em\u003e, consistent with enhancer\u0026ndash;promoter activation in epigenomic and single-cell transcriptomic data. We also identified AD-associated deletions at the \u003cem\u003eKANSL1\u003c/em\u003e and \u003cem\u003eARMS2\u003c/em\u003e that rewire local regulatory contacts and associate with cell-type-specific expression changes, demonstrating \u003cem\u003eiNeuroAkita\u0026rsquo;s\u003c/em\u003e utility for systematic characterization of noncoding AD risk variants underlying 3D genome regulation.\u003c/p\u003e","manuscriptTitle":"Neuronal and microglial fine-tuned 3D DNA foundation models dissect genetic variants in Alzheimer’s disease","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-26 17:12:24","doi":"10.21203/rs.3.rs-8920766/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":"ad3f5c77-38bb-4bc9-b568-2566c61f1bf3","owner":[],"postedDate":"February 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":63535313,"name":"Biological sciences/Genetics/Gene regulation"},{"id":63535314,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":63535315,"name":"Health sciences/Neurology/Neurological disorders/Dementia/Alzheimer's disease"}],"tags":[],"updatedAt":"2026-03-27T12:12:01+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-26 17:12:24","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8920766","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8920766","identity":"rs-8920766","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}