Protein Set Transformer: A protein-based genome language model to power high diversity viromics

Protein Set Transformer: A protein-based genome language model to power high diversity viromics | 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 Protein Set Transformer: A protein-based genome language model to power high diversity viromics Karthik Anantharaman, Cody Martin, Anthony Gitter This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4844047/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Nov, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Exponential increases in microbial and viral genomic data demand transformational advances in scalable, generalizable frameworks for their interpretation. Standard homology-based functional analyses are hindered by the rapid divergence of microbial and especially viral genomes and proteins that significantly decreases the volume of usable data. Here, we present Protein Set Transformer (PST), a protein-based genome language model that models genomes as sets of proteins without considering sparsely available functional labels. Trained on >100k viruses, PST outperformed other homology- and language model-based approaches for relating viral genomes based on shared protein content. Further, PST demonstrated protein structural and functional awareness by clustering capsid-fold-containing proteins with known capsid proteins and uniquely clustering late gene proteins within related viruses. Our data establish PST as a valuable method for diverse viral genomics, ecology, and evolutionary applications. We posit that the PST framework can be a foundation model for microbial genomics when trained on suitable data. Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Microbiology/Virology/Metagenomics viromics viral metagenomics genome language model protein language model transformer Full Text Additional Declarations There is NO Competing Interest. Supplementary Files ExtendedDataFigures.docx Cite Share Download PDF Status: Published Journal Publication published 23 Nov, 2025 Read the published version in Nature Communications → 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-4844047","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":353184806,"identity":"54ae1866-ada2-45e7-afe4-5106f42590b4","order_by":0,"name":"Karthik Anantharaman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYBACCSBmZjBIkDNA8GEkbi2MzUAtxkAtjA0kaGFISNwA0YJsFw4g2X72+eOCgrT07eyHnz9gbLOw529gPnibB48WaZ50w+YZBjm5O3vSDBsY2yQSZxxgS7bGp0WOIY2xmcegInfDgRzGBsZtEgkGDDxm0ni18D8Da0k3OP8GrMXegIH/G14t0hJgW3ISDG5AbGHcwMDDhleL5IxnjLN5DNIMd854Zjgj8R/QL4fZjC3n4NEicT6N4TPPn2R5c/7kBx8+nKmz529vfnjjDR4tqCABRDATrXwUjIJRMApGAS4AAOosRIi/xeGTAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-9584-2491","institution":"University of Wisconsin-Madison","correspondingAuthor":true,"prefix":"","firstName":"Karthik","middleName":"","lastName":"Anantharaman","suffix":""},{"id":353184807,"identity":"3ed81947-0abf-4e64-b01e-9f7b287155f6","order_by":1,"name":"Cody Martin","email":"","orcid":"https://orcid.org/0000-0001-7610-9537","institution":"University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Cody","middleName":"","lastName":"Martin","suffix":""},{"id":353184808,"identity":"615e9d4b-95db-42c2-90ad-e00893948a20","order_by":2,"name":"Anthony Gitter","email":"","orcid":"https://orcid.org/0000-0002-5324-9833","institution":"University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Anthony","middleName":"","lastName":"Gitter","suffix":""}],"badges":[],"createdAt":"2024-08-01 18:15:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4844047/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4844047/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-66049-4","type":"published","date":"2025-11-23T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":98290931,"identity":"8b06473f-843d-4129-85f2-4115d301429d","added_by":"auto","created_at":"2025-12-16 08:11:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4220244,"visible":true,"origin":"","legend":"","description":"","filename":"pstmanuscriptWITHFIGURES.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4844047/v1_covered_a2c62d56-86ef-43da-95a9-2e1d4077049e.pdf"},{"id":65037887,"identity":"5aa22528-9755-452d-92b8-db298398e1f2","added_by":"auto","created_at":"2024-09-23 02:26:35","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4637128,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-4844047/v1/8a927d5802512a8381feace1.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Protein Set Transformer: A protein-based genome language model to power high diversity viromics","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"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":"viromics, viral metagenomics, genome language model, protein language model, transformer","lastPublishedDoi":"10.21203/rs.3.rs-4844047/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4844047/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Exponential increases in microbial and viral genomic data demand transformational advances in scalable, generalizable frameworks for their interpretation. Standard homology-based functional analyses are hindered by the rapid divergence of microbial and especially viral genomes and proteins that significantly decreases the volume of usable data. Here, we present Protein Set Transformer (PST), a protein-based genome language model that models genomes as sets of proteins without considering sparsely available functional labels. Trained on \u003e100k viruses, PST outperformed other homology- and language model-based approaches for relating viral genomes based on shared protein content. Further, PST demonstrated protein structural and functional awareness by clustering capsid-fold-containing proteins with known capsid proteins and uniquely clustering late gene proteins within related viruses. Our data establish PST as a valuable method for diverse viral genomics, ecology, and evolutionary applications. We posit that the PST framework can be a foundation model for microbial genomics when trained on suitable data.","manuscriptTitle":"Protein Set Transformer: A protein-based genome language model to power high diversity viromics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-23 02:26:31","doi":"10.21203/rs.3.rs-4844047/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":"8e460b50-715b-4e8c-91f7-8ffc5f1e2c7e","owner":[],"postedDate":"September 23rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":37522485,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":37522486,"name":"Biological sciences/Microbiology/Virology/Metagenomics"}],"tags":[],"updatedAt":"2025-12-16T08:10:38+00:00","versionOfRecord":{"articleIdentity":"rs-4844047","link":"https://doi.org/10.1038/s41467-025-66049-4","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2025-11-23 05:00:00","publishedOnDateReadable":"November 23rd, 2025"},"versionCreatedAt":"2024-09-23 02:26:31","video":"","vorDoi":"10.1038/s41467-025-66049-4","vorDoiUrl":"https://doi.org/10.1038/s41467-025-66049-4","workflowStages":[]},"version":"v1","identity":"rs-4844047","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4844047","identity":"rs-4844047","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}