OmegaFold: Deep Learning Paradigm for Universal Protein Structure Prediction via Attention-Based Geometric Transformers and Evolutionary Language Modeling

OmegaFold: Deep Learning Paradigm for Universal Protein Structure Prediction via Attention-Based Geometric Transformers and Evolutionary Language Modeling | 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 OmegaFold: Deep Learning Paradigm for Universal Protein Structure Prediction via Attention-Based Geometric Transformers and Evolutionary Language Modeling KalyanChakravarthy Kodela This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7514283/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 This paper introduces a groundbreaking computational paradigm for protein structure prediction through novel OmegaFold architecture that fundamentally transforms traditional multiple sequence alignment dependent methodologies. Theresearch establishes an unprecedented theoretical framework incorporating transformer-based attention mechanisms, geometric deep learning principles, and evolutionary language modeling to achieve universal folding prediction capabilities. The methodology demonstrates exceptional performance metrics including Template Modeling scores exceeding 0.85 for diverse protein families while maintaining computational efficiency superior to conventional approaches. Experimental validation across comprehensive benchmarks reveals remarkable accuracy improvements of approximately 15-20 percent compared to existing state-of-the-art methods, establishing new performance standards for structural biology applications. Artificial Intelligence Transformer Architecture Geometric Deep Learning Protein Folding Computational Biology Machine Learning Structural Genomics Bioinformatics Neural Networks Attention Mechanisms Full Text Additional Declarations The authors declare no competing interests. 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-7514283","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":508899648,"identity":"d5ae04cb-1175-4b63-b2ec-37f8b4bc715f","order_by":0,"name":"KalyanChakravarthy Kodela","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABMUlEQVRIie2RMWuEMBSAnwh2CedqkN5viBUOj/bHJAhO3u50KIK39G4+aX9Ef4LiYAfpLHS4c7+hpVCEOjTmKG0h0rVDviGEl/cl770AKBT/EPNiUx4pUIBCj+FlDBXiwPqx/gbfNj45CkWLtT2QUdFinmxNKaSlC/yl6OhbgUkF72lgUQjmszpJ3m6Gge3qbd314K1JoZfPSNIL6iqPQujipkztVUZY3jyxFPHCSGH41xIFbyhrKUTsoWWxvYoJ34ROCkJBC1uikIISSyiHLv3wBq4cTk7SC8V8lyqP9IorIb9cy2wwxleQE58LQ4ZMEUOmJOC9sGy5zVw3b0LnDhEL55XhLu8l7fOv7PrI5xOrqrYf5pe7uiGvfbQ2Z3XatSfJlM/lSSP6VLpCoVAo/uIT1ShxWzSNOgkAAAAASUVORK5CYII=","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"KalyanChakravarthy","middleName":"","lastName":"Kodela","suffix":""}],"badges":[],"createdAt":"2025-09-02 06:57:03","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-7514283/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7514283/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90493882,"identity":"3f0b036e-38c4-4a03-aca1-25cbda08d7bd","added_by":"auto","created_at":"2025-09-03 10:12:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":503234,"visible":true,"origin":"","legend":"","description":"","filename":"omegafoldresearch.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7514283/v1_covered_d1173138-f22d-47d2-8c41-f675d9e02aa0.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eOmegaFold: Deep Learning Paradigm for Universal Protein Structure Prediction via Attention-Based Geometric Transformers and Evolutionary Language Modeling\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"International Technological University","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":"Artificial Intelligence, Transformer Architecture, Geometric Deep Learning, Protein Folding, Computational Biology, Machine Learning, Structural Genomics, Bioinformatics, Neural Networks, Attention Mechanisms","lastPublishedDoi":"10.21203/rs.3.rs-7514283/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7514283/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper introduces a groundbreaking computational paradigm for protein structure prediction through novel OmegaFold architecture that fundamentally transforms traditional multiple sequence alignment dependent methodologies. Theresearch establishes an unprecedented theoretical framework incorporating transformer-based attention mechanisms, geometric deep learning principles, and evolutionary language modeling to achieve universal folding prediction capabilities. The methodology demonstrates exceptional performance metrics including Template Modeling scores exceeding 0.85 for diverse protein families while maintaining computational efficiency superior to conventional approaches. Experimental validation across comprehensive benchmarks reveals remarkable accuracy improvements of approximately 15-20 percent compared to existing state-of-the-art methods, establishing new performance standards for structural biology applications.\u003c/p\u003e","manuscriptTitle":"OmegaFold: Deep Learning Paradigm for Universal Protein Structure Prediction via Attention-Based Geometric Transformers and Evolutionary Language Modeling","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-03 10:04:42","doi":"10.21203/rs.3.rs-7514283/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":"fdb1a126-4082-4069-a356-dfd0de820e7c","owner":[],"postedDate":"September 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-09-03T10:04:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-03 10:04:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7514283","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7514283","identity":"rs-7514283","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}