{"paper_id":"35fc944a-eb8f-4d1a-ac70-58b6037ac059","body_text":"Precise and interpretable neural networks reveal epigenetic signatures of aging in youth across health and 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 Research Article Precise and interpretable neural networks reveal epigenetic signatures of aging in youth across health and disease David Martínez-Enguita, Thomas Hillerton, Julia Åkesson, Daniel Kling, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4863132/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 DNA methylation age clocks are powerful tools for measuring biological age, providing insights into aging risks and outcomes beyond chronological age. While traditional clocks are effective, their interpretability is often limited due to their dependence on small, potentially stochastic sets of CpG sites. We propose that the reliability of epigenetic age estimation models should stem from their ability to detect comprehensive and targeted aging signatures. Here we introduce NCAE-CombClock, a neural network regressor that improves age prediction accuracy in large datasets by integrating methylation embeddings with CpG sites. Additionally, we developed explainable neural networks for robust age classification across adolescence and young adulthood. Epigenetic aging signatures identified at single-year resolution from these models were enriched in developmental, immune, and metabolic processes. We showcase the utility of our approach by exploring the biological mechanisms underlying the altered pace of aging observed in pediatric Crohn's disease. Our models offer broad applications in personalized medicine and aging research, providing a valuable resource for interpreting aging trajectories in health and disease. Bioinformatics Epigenetics & Genomics DNA methylation neural networks age clock epigenetic age aging youth Full Text Additional Declarations The authors declare potential competing interests as follows: DM-E, ML, and MG are founders of PredictMe, a company that provides DNA methylation analysis services. The other authors declare no competing interests Supplementary Files suppltablefigures.pdf Supplementary Tables and Figures supplfile1.xlsx Supplementary File 1: Summary of DNA methylation sample sets for DNAm age clocks training and evaluation. supplfile2.xlsx Supplementary File 2: CpG sites and coefficients of benchmarked DNAm age clocks. supplfile3.xlsx Supplementary File 3: Regression and classification benchmark performance of DNAm age clocks. supplfile4.xlsx Supplementary File 4: Estimated probabilities from NCAE-Age classifier array. supplfile5.xlsx Supplementary File 5: DNAm signatures of aging from NCAE-Age clock and classifier array. supplfile6.xlsx Supplementary File 6: DNAm signatures of aging from NCAE-XY-Age clock and classifier array. supplfile7.xlsx Supplementary File 7: Functional enrichment analyses of CpG sets and DNAm signatures of aging from DNAm age clocks and NCAE-Age classifier arrays. 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-4863132\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":336264017,\"identity\":\"5bae4289-7d2a-4dd8-b2ca-7a17ca1e0ad3\",\"order_by\":0,\"name\":\"David Martínez-Enguita\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-6363-6298\",\"institution\":\"Linköping University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"David\",\"middleName\":\"\",\"lastName\":\"Martínez-Enguita\",\"suffix\":\"\"},{\"id\":336268813,\"identity\":\"3b011f0a-889b-483e-b535-5e9df6a74687\",\"order_by\":1,\"name\":\"Thomas Hillerton\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Linköping University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Thomas\",\"middleName\":\"\",\"lastName\":\"Hillerton\",\"suffix\":\"\"},{\"id\":336268814,\"identity\":\"95e4a764-cdf1-499f-8ebd-aafb5e8170b3\",\"order_by\":2,\"name\":\"Julia Åkesson\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Linköping University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Julia\",\"middleName\":\"\",\"lastName\":\"Åkesson\",\"suffix\":\"\"},{\"id\":336268815,\"identity\":\"50a82850-a562-4c2f-ba28-d4ffd8fbe6c6\",\"order_by\":3,\"name\":\"Daniel Kling\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Oslo University Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Daniel\",\"middleName\":\"\",\"lastName\":\"Kling\",\"suffix\":\"\"},{\"id\":336268816,\"identity\":\"48770724-7526-45d0-a2f0-38688ae44183\",\"order_by\":4,\"name\":\"Maria Lerm\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Linköping University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Maria\",\"middleName\":\"\",\"lastName\":\"Lerm\",\"suffix\":\"\"},{\"id\":336268817,\"identity\":\"77f6e2ba-1340-40d5-969c-cf9ea8f6ebee\",\"order_by\":5,\"name\":\"Mika Gustafsson\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYDACZjYYi/EBiJTjYyZWCw8DswGINmYjqIUBTUtiGyENuu1siR8Yfh22t2c/zPzxR8Wd9DZ2HgPGHxW4tZgdZjsswdh3OLGHJ5lNmufMs9w2Zh4DZp4z+LSwN0gw9hxO4GHIP8bM2HYYooURj/OAWpp/ALXY8/A/Zv74s+1wOhtQC+PPf3gddkyC4cdhxh6JZAYJ3rbDCSAtDLwNeLWkWSQ2pCf23HgM8sthwzZmtoLDPMfwaDl/zPjGhz/W9uz9yaAQOyzPz39448MfNbi1gAFGXBwgoAEI/hBWMgpGwSgYBSMYAABZaUn2N+t9JgAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"Linköping University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Mika\",\"middleName\":\"\",\"lastName\":\"Gustafsson\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-08-05 15:38:52\",\"currentVersionCode\":1,\"declarations\":{\"humanSubjects\":false,\"vertebrateSubjects\":false,\"conflictsOfInterestStatement\":true,\"humanSubjectEthicalGuidelines\":false,\"humanSubjectConsent\":false,\"humanSubjectClinicalTrial\":false,\"humanSubjectCaseReport\":false,\"vertebrateSubjectEthicalGuidelines\":false},\"doi\":\"10.21203/rs.3.rs-4863132/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-4863132/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":61911267,\"identity\":\"b5b9da4d-9b80-4b9e-8a43-a90064c0a96e\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:19:22\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":6442776,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1_covered_d5fe7f6f-bb1c-4c87-ad0d-3bb3e54f69ed.pdf\"},{\"id\":61910502,\"identity\":\"059c11cd-9b85-4cde-b6e6-eb46be4c0ea3\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:13\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":993202,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary Tables and Figures\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"suppltablefigures.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/2a26a84449c10e466de86e9e.pdf\"},{\"id\":61910918,\"identity\":\"6a8656c5-f2f9-4f42-9f4e-1ed0a8c6720f\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:11:13\",\"extension\":\"xlsx\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":2169136,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 1: Summary of DNA methylation sample sets for DNAm age clocks training and evaluation.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile1.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/b0c600a812d2675332cdfeb8.xlsx\"},{\"id\":61910504,\"identity\":\"79d9226b-5aca-4497-8cd2-59cd51930590\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:13\",\"extension\":\"xlsx\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":12558675,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 2: CpG sites and coefficients of benchmarked DNAm age clocks.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile2.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/73eb431291de6e3e58883186.xlsx\"},{\"id\":61910503,\"identity\":\"bdb36218-3acf-49a2-a032-d5539b155a9e\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:13\",\"extension\":\"xlsx\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":382031,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 3: Regression and classification benchmark performance of DNAm age clocks.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile3.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/f1fdd898467d23d11a431fe1.xlsx\"},{\"id\":61910501,\"identity\":\"d7ed1f5a-4c7f-4622-ada6-02db62407fd9\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:13\",\"extension\":\"xlsx\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":111029,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 4: Estimated probabilities from NCAE-Age classifier array.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile4.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/d742a25beb1f2fd8ed88742e.xlsx\"},{\"id\":61910507,\"identity\":\"b59296e0-9d69-4dbc-8ec6-60aa048560e0\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:14\",\"extension\":\"xlsx\",\"order_by\":6,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":54399069,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 5: DNAm signatures of aging from NCAE-Age clock and classifier\\u003c/p\\u003e\\n\\u003cp\\u003earray.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile5.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/2aef2d4a8a0453d4883a4402.xlsx\"},{\"id\":61910508,\"identity\":\"84c33b60-7447-48b6-90bd-2e312348bd04\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:14\",\"extension\":\"xlsx\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":55309833,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 6: DNAm signatures of aging from NCAE-XY-Age clock and\\u003c/p\\u003e\\n\\u003cp\\u003eclassifier array.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile6.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/cfe8b08504e881a39be395ce.xlsx\"},{\"id\":61910506,\"identity\":\"c9212035-45b3-480a-8605-08c03f8d1c4c\",\"added_by\":\"auto\",\"created_at\":\"2024-08-07 03:03:13\",\"extension\":\"xlsx\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":456506,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSupplementary File 7: Functional enrichment analyses of CpG sets and DNAm\\u003c/p\\u003e\\n\\u003cp\\u003esignatures of aging from DNAm age clocks and NCAE-Age classifier arrays.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"supplfile7.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4863132/v1/760af623d2070edb95d94880.xlsx\"}],\"financialInterests\":\"The authors declare potential competing interests as follows: DM-E, ML, and MG are founders of PredictMe, a company that provides DNA methylation analysis services. The other authors declare no competing interests\",\"formattedTitle\":\"\\u003cp\\u003ePrecise and interpretable neural networks reveal epigenetic signatures of aging in youth across health and disease\\u003c/p\\u003e\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"Linköping University\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":true,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":true,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"DNA methylation, neural networks, age clock, epigenetic age, aging, youth\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-4863132/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-4863132/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003eDNA methylation age clocks are powerful tools for measuring biological age, providing insights into aging risks and outcomes beyond chronological age. While traditional clocks are effective, their interpretability is often limited due to their dependence on small, potentially stochastic sets of CpG sites. We propose that the reliability of epigenetic age estimation models should stem from their ability to detect comprehensive and targeted aging signatures. Here we introduce NCAE-CombClock, a neural network regressor that improves age prediction accuracy in large datasets by integrating methylation embeddings with CpG sites. Additionally, we developed explainable neural networks for robust age classification across adolescence and young adulthood. Epigenetic aging signatures identified at single-year resolution from these models were enriched in developmental, immune, and metabolic processes. We showcase the utility of our approach by exploring the biological mechanisms underlying the altered pace of aging observed in pediatric Crohn's disease. Our models offer broad applications in personalized medicine and aging research, providing a valuable resource for interpreting aging trajectories in health and disease.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Precise and interpretable neural networks reveal epigenetic signatures of aging in youth across health and disease\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-08-07 03:03:08\",\"doi\":\"10.21203/rs.3.rs-4863132/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"438c407a-de31-4c8d-ba99-513923e747fb\",\"owner\":[],\"postedDate\":\"August 7th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[{\"id\":35615699,\"name\":\"Bioinformatics\"},{\"id\":35615700,\"name\":\"Epigenetics \\u0026 Genomics\"}],\"tags\":[],\"updatedAt\":\"2024-08-07T03:03:08+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-08-07 03:03:08\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-4863132\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-4863132\",\"identity\":\"rs-4863132\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}