RAGP: a retrieval-augmented deep learning model for genomic prediction in crop breeding

RAGP: a retrieval-augmented deep learning model for genomic prediction in crop breeding | 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 RAGP: a retrieval-augmented deep learning model for genomic prediction in crop breeding Lingling Zuo, Xinger Li, Xiaoli Wang, Rui Man, Yang Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7229420/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Genomic prediction (GP) plays a pivotal role in expediting genetic gains for crop breeding. Recently, deep learning models has attracted growing attention in this field due to its superior performance over traditional models. However, most advanced models rely on simplified numerical representations of genomic variants and treat individuals in isolation, failing to capture the full complexity of genetic variation and the genetic relatedness among individuals. In response, we propose RAGP, a deep learning model driven by a retrieval-augmented mechanism composed of two synergistic components: 1) an embedding-based retrieval module that identifies a group of nearest neighbors highly relevant to the target sample as references, employs a gene-specific discriminator to refine the representation space and capture informative genetic structures; 2) an augmentation module that integrates the retrieved references to enhance the target representation. The enriched representation is then propagated through a regression network for final prediction. This retrieval-guided strategy enables more informed and fine-grained utilization of genomic data and strengthens the model’s capacity to capture individual-level genetic relationships. Extensive experiments reveal that RAGP achieves consistently superior predictive performance compared to both conventional methods and state-of-the-art deep learning models. By incorporating retrieval-augmented mechanisms, RAGP achieves substantial improvements in predictive accuracy and offers a promising direction for advancing genomic selection in crop breeding. Our code is available on https://github.com/l00907l/RAGP . deep learning genomic prediction retrieval-augmentation crop breeding Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 29 Dec, 2025 Reviews received at journal 22 Dec, 2025 Reviewers agreed at journal 17 Dec, 2025 Reviews received at journal 03 Dec, 2025 Reviewers agreed at journal 06 Nov, 2025 Reviewers invited by journal 31 Oct, 2025 Editor assigned by journal 29 Jul, 2025 Submission checks completed at journal 29 Jul, 2025 First submitted to journal 27 Jul, 2025 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-7229420","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":542602702,"identity":"4db61afd-2b67-45b5-8221-e4bdd1dc995e","order_by":0,"name":"Lingling Zuo","email":"","orcid":"","institution":"Nanjing University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Lingling","middleName":"","lastName":"Zuo","suffix":""},{"id":542602704,"identity":"5194672b-4b0a-4744-9d2c-1378efb2f3c4","order_by":1,"name":"Xinger Li","email":"","orcid":"","institution":"Nanjing University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xinger","middleName":"","lastName":"Li","suffix":""},{"id":542602705,"identity":"c634ea38-78af-4ca5-b7aa-3ecb3beefc8a","order_by":2,"name":"Xiaoli Wang","email":"","orcid":"","institution":"Agricultural Information Institute of CAAS","correspondingAuthor":false,"prefix":"","firstName":"Xiaoli","middleName":"","lastName":"Wang","suffix":""},{"id":542602706,"identity":"b2a5c289-007b-4db1-be26-c0552cf479f5","order_by":3,"name":"Rui Man","email":"","orcid":"","institution":"Agricultural Information Institute of CAAS","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Man","suffix":""},{"id":542602707,"identity":"2d494697-5209-402d-beda-96ffe1868bf9","order_by":4,"name":"Yang Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIie3LLwvCQBjH8d94QMvUumP45yUMBA2Cr2Vw4Ip5GAwmk2AVFF/G8smBFnHGBYMWk0GDsiJ4w+7NJnhfeB7u4PkAJtPPNgC9H4XcZJsR+oZY42znJV7Cz6f78lD04ljgGkpU5iMd6bWb1ehMysKa7SScg9ARv+WySCpCoNJYwnN8HQkeLlsoEkvQMx/pt9htpIjgICsPYdtL6GItiSXcW012ge0kGlLeBBFLh5KX49XpmIadWmWmIY29D7IBnn2EGvvzvao+FbBSoKu9NJlMpv/tBUhYRe8Ck7VmAAAAAElFTkSuQmCC","orcid":"","institution":"Nanjing University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-07-28 03:38:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7229420/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7229420/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95662451,"identity":"db362865-5969-493c-b686-94ebe43e89ac","added_by":"auto","created_at":"2025-11-11 16:37:36","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3233175,"visible":true,"origin":"","legend":"","description":"","filename":"mianmanuscriptofRAGP.docx","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/526e37b0da183a87b44d7b6a.docx"},{"id":95662394,"identity":"5dc71acc-7122-4948-9aa5-187f1d3bfca7","added_by":"auto","created_at":"2025-11-11 16:37:29","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6642,"visible":true,"origin":"","legend":"","description":"","filename":"84612a3a29cc4d36afed4d0663af9027.json","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/82d95e141024a03ab4ea4d80.json"},{"id":95662459,"identity":"15869dfd-16f8-465c-bb26-e980d93995fc","added_by":"auto","created_at":"2025-11-11 16:37:37","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115354,"visible":true,"origin":"","legend":"","description":"","filename":"84612a3a29cc4d36afed4d0663af90271enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/d81ec4391b3b07a04acedf22.xml"},{"id":95662567,"identity":"c4b88131-feda-4339-a8f1-ed9f043640e6","added_by":"auto","created_at":"2025-11-11 16:37:43","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":318590,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/18b5899a6970dc92ae4e0034.png"},{"id":95662569,"identity":"d225e93d-6ce7-4c74-b15f-ff7ce536a19e","added_by":"auto","created_at":"2025-11-11 16:37:43","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76530,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/e77a48e512cb4420657e083d.png"},{"id":95662470,"identity":"3e6d0f35-e50d-414f-b55f-0c5da464df7c","added_by":"auto","created_at":"2025-11-11 16:37:38","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78954,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/e656b50e18ba2327873f2e1f.png"},{"id":95662433,"identity":"478f625d-711d-44c0-991a-1d95fcf2c233","added_by":"auto","created_at":"2025-11-11 16:37:34","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55096,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/606114013d305127e84f1ef5.png"},{"id":95662474,"identity":"bd4b5146-2547-4bd0-b7a7-c33a2533a383","added_by":"auto","created_at":"2025-11-11 16:37:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":73923,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/f8bd4bf0d3e8de6b26a736a0.png"},{"id":95662434,"identity":"59a4c0af-62a5-4d9a-9060-ddc35a59a252","added_by":"auto","created_at":"2025-11-11 16:37:34","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28268,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/5b09543b02851c51198506a5.png"},{"id":95662473,"identity":"969a6653-74b0-493b-89c2-1901145b8780","added_by":"auto","created_at":"2025-11-11 16:37:38","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":187990,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/677294f85264b54dee180ad3.png"},{"id":95662430,"identity":"2c956899-1326-48c3-9332-3e25ef1e3360","added_by":"auto","created_at":"2025-11-11 16:37:33","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114488,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/4450fa3219bcc1337fe795e8.png"},{"id":95662475,"identity":"805638fc-4c5c-46c5-beb5-e79b1babbee0","added_by":"auto","created_at":"2025-11-11 16:37:39","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18352,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/339119d6f9d387c153ee6046.png"},{"id":95662441,"identity":"f59923d8-8644-4e5c-aeaa-060f48295145","added_by":"auto","created_at":"2025-11-11 16:37:35","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20233,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/03daa182448507b066d5dbb0.png"},{"id":95662477,"identity":"c854801f-5e26-48ab-b594-5ad941e79003","added_by":"auto","created_at":"2025-11-11 16:37:39","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12200,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/f7d79482e654b7cb4d87e8a9.png"},{"id":95662447,"identity":"ad0f789b-3fea-40c0-93b5-351c8f7db75a","added_by":"auto","created_at":"2025-11-11 16:37:36","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19455,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/9a3178bb19fc1a8911d82f12.png"},{"id":95662542,"identity":"d4767445-13b5-4a8c-9082-fedb17107121","added_by":"auto","created_at":"2025-11-11 16:37:42","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10064,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/36b30c9670b57ccb64a3da7d.png"},{"id":95662571,"identity":"0a27df6c-57f1-40aa-80b8-b796d5311edd","added_by":"auto","created_at":"2025-11-11 16:37:44","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57306,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/fa9878317d6b697ca6c6e897.png"},{"id":95662440,"identity":"0a7a0cb9-923c-45b9-8be9-3ca01f09e51c","added_by":"auto","created_at":"2025-11-11 16:37:35","extension":"xml","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":113948,"visible":true,"origin":"","legend":"","description":"","filename":"84612a3a29cc4d36afed4d0663af90271structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/ac61324087fbc70c6db5f627.xml"},{"id":95662465,"identity":"c83015f7-acaf-402a-b26c-e3d0673db3e4","added_by":"auto","created_at":"2025-11-11 16:37:37","extension":"html","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125830,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1/cd05d83a8fd138064baceffc.html"},{"id":95663584,"identity":"c613b98a-9203-40cc-95c0-9cabe2c7bc4b","added_by":"auto","created_at":"2025-11-11 16:39:07","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":977122,"visible":true,"origin":"","legend":"","description":"","filename":"mianmanuscriptofRAGP.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7229420/v1_covered_2db3e124-29e6-4852-b0a2-8660e62b61b4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"RAGP: a retrieval-augmented deep learning model for genomic prediction in crop breeding","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"plant-methods","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"plme","sideBox":"Learn more about [Plant Methods](http://plantmethods.biomedcentral.com/)","snPcode":"13007","submissionUrl":"https://submission.nature.com/new-submission/13007/3","title":"Plant Methods","twitterHandle":"@PlantMethods","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"deep learning, genomic prediction, retrieval-augmentation, crop breeding","lastPublishedDoi":"10.21203/rs.3.rs-7229420/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7229420/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGenomic prediction (GP) plays a pivotal role in expediting genetic gains for crop breeding. Recently, deep learning models has attracted growing attention in this field due to its superior performance over traditional models. However, most advanced models rely on simplified numerical representations of genomic variants and treat individuals in isolation, failing to capture the full complexity of genetic variation and the genetic relatedness among individuals. In response, we propose RAGP, a deep learning model driven by a retrieval-augmented mechanism composed of two synergistic components: 1) an embedding-based retrieval module that identifies a group of nearest neighbors highly relevant to the target sample as references, employs a gene-specific discriminator to refine the representation space and capture informative genetic structures; 2) an augmentation module that integrates the retrieved references to enhance the target representation. The enriched representation is then propagated through a regression network for final prediction. This retrieval-guided strategy enables more informed and fine-grained utilization of genomic data and strengthens the model\u0026rsquo;s capacity to capture individual-level genetic relationships. Extensive experiments reveal that RAGP achieves consistently superior predictive performance compared to both conventional methods and state-of-the-art deep learning models. By incorporating retrieval-augmented mechanisms, RAGP achieves substantial improvements in predictive accuracy and offers a promising direction for advancing genomic selection in crop breeding. Our code is available on \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/l00907l/RAGP\u003c/span\u003e\u003cspan address=\"https://github.com/l00907l/RAGP\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e","manuscriptTitle":"RAGP: a retrieval-augmented deep learning model for genomic prediction in crop breeding","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 16:23:50","doi":"10.21203/rs.3.rs-7229420/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-29T22:09:27+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-22T17:37:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"198145565452339414642486909150653322621","date":"2025-12-17T08:37:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-04T04:48:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"265020616252024822302390785737194001287","date":"2025-11-06T18:59:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-31T10:34:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-29T06:25:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-29T06:24:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"Plant Methods","date":"2025-07-28T03:34:03+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"plant-methods","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"plme","sideBox":"Learn more about [Plant Methods](http://plantmethods.biomedcentral.com/)","snPcode":"13007","submissionUrl":"https://submission.nature.com/new-submission/13007/3","title":"Plant Methods","twitterHandle":"@PlantMethods","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b5a1a115-1a9b-484a-aa5e-dc5ecc0aabfc","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-15T04:23:20+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 16:23:50","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7229420","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7229420","identity":"rs-7229420","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}