Hierarchical Clustering-Based Coarse-to-Fine Classification Framework for Microbial Protein Function Prediction

Hierarchical Clustering-Based Coarse-to-Fine Classification Framework for Microbial Protein Function Prediction | 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 Hierarchical Clustering-Based Coarse-to-Fine Classification Framework for Microbial Protein Function Prediction Shengyang Chen, Xinyue Gao, Congmin Zhu, Honglei Liu, Yuqing Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7473073/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Dec, 2025 Read the published version in BMC Bioinformatics → Version 1 posted 12 You are reading this latest preprint version Abstract Background: Accurate prediction of microbial protein functions is essential for understanding microbial physiology, discovering novel probiotics, and driving biotechnological innovation. However, protein function prediction remains challenging due to the hierarchical and class-imbalanced nature of functional labels, particularly in large-scale annotations such as Enzyme Commission (EC) numbers and Gene Ontology (GO) terms. Most existing deep learning approaches fail to adequately address the long-tail distribution problem. Methods: We propose a Hierarchical Cascaded Context Network (HCCN) that explicitly models functional hierarchies and emphasizes prediction of low-frequency (long-tail) labels. For EC classification, we design a coarse-to-fine network that captures parent–child dependencies among hierarchical labels. For GO prediction, we construct a semantically grounded hierarchical structure using ontology embedding and clustering, and develop an attention-based multi-level cascade predictor to exploit structured dependencies across Biological Process (BPO), Molecular Function (MFO), and Cellular Component (CCO). To mitigate label imbalance, we introduce a dynamic resampling strategy and a hierarchical loss weighting mechanism, which enforce inter-level regularization and enhance sensitivity to rare functions. Results: Experimental results show that HCCN consistently outperforms traditional sequence-alignment methods (e.g., DIAMOND, BLAST) and baseline neural networks (MLP) across all major functional categories. On the full test set, HCCN achieves AUPR gains of up to 9.3% (EC), 10.6% (BPO), 6.9% (MFO), and 5.4% (CCO) over the best baseline. For low-frequency labels, HCCN demonstrates strong few-shot generalization, with improvements of + 11.2% (EC low) and + 10.9% (BPO low) in mAUPR. Conclusions: The proposed HCCN framework provides an effective solution to hierarchical and imbalanced protein function prediction, significantly improving performance on long-tail functional labels. Code and data are publicly available at: https://github.com/YangLab-BUPT/HCCN . Protein Function Prediction Deep learning Gene Ontology Coarse-to-Fine Classification Protein Sequence Long-tail Label Prediction Full Text Additional Declarations No competing interests reported. Supplementary Files TableS1.xlsx TableS2.xlsx Cite Share Download PDF Status: Published Journal Publication published 20 Dec, 2025 Read the published version in BMC Bioinformatics → Version 1 posted Editorial decision: Revision requested 09 Oct, 2025 Reviews received at journal 04 Oct, 2025 Reviewers agreed at journal 23 Sep, 2025 Reviewers agreed at journal 21 Sep, 2025 Reviewers agreed at journal 20 Sep, 2025 Reviews received at journal 05 Sep, 2025 Reviewers agreed at journal 04 Sep, 2025 Reviewers invited by journal 04 Sep, 2025 Editor invited by journal 04 Sep, 2025 Editor assigned by journal 27 Aug, 2025 Submission checks completed at journal 27 Aug, 2025 First submitted to journal 27 Aug, 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-7473073","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":512817176,"identity":"fa914b92-12ab-48fe-8791-6eecb7ca7c80","order_by":0,"name":"Shengyang Chen","email":"","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Shengyang","middleName":"","lastName":"Chen","suffix":""},{"id":512817177,"identity":"fc5b4c65-040b-4338-8878-77c287fba09b","order_by":1,"name":"Xinyue Gao","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Xinyue","middleName":"","lastName":"Gao","suffix":""},{"id":512817178,"identity":"9460a2ea-fc1a-4d8f-a94a-ca1f3010a863","order_by":2,"name":"Congmin Zhu","email":"","orcid":"","institution":"Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Congmin","middleName":"","lastName":"Zhu","suffix":""},{"id":512817179,"identity":"acafdc84-06c7-478b-9944-3429db9a878e","order_by":3,"name":"Honglei Liu","email":"","orcid":"","institution":"Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Honglei","middleName":"","lastName":"Liu","suffix":""},{"id":512817180,"identity":"504dfac9-0c18-4d72-973b-62f44e24de0f","order_by":4,"name":"Yuqing Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYBADHgYG5mNQdgLRWtjSgLQB8VpAusyI02Jw/Ozh17xt1jLm/Gu+Pebd8YeBnz3HgOHnDjxazuSlWfO2pfNYzni73Zj3jAGDZM8bA8beM3i0HMgxM+ZtO8xjcOPsNmneNgMGgxs5BsyMbXi0nH8D03LmGViLPUEtN3KMH4O1nO9hg9giQUCL5I03ZoxzzqUDbWEzN5x7xphH4syzgoO9eLTwnc8x/vCmzNre4PzhZw/e7pCT429P3vjgJx4tCgcY2KR4GJgZGCQSGBgYG0BxysBwALcGBgb5Bgbmjz9AWvgPgLWMglEwCkbBKMAAAGehUdPaHTBBAAAAAElFTkSuQmCC","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":true,"prefix":"","firstName":"Yuqing","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-08-27 15:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7473073/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7473073/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12859-025-06326-7","type":"published","date":"2025-12-20T15:57:44+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":98813917,"identity":"c52baeba-7272-4b1e-a7c9-30e2e576d5f0","added_by":"auto","created_at":"2025-12-22 16:07:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":909119,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript25825.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7473073/v1_covered_10df6a3d-65fa-4e12-be7b-f26c7ce7fd6f.pdf"},{"id":91102365,"identity":"c7d827aa-baf6-474c-91f3-3deb176be185","added_by":"auto","created_at":"2025-09-11 14:57:37","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8984,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7473073/v1/a6f43a892470f353dac0eb20.xlsx"},{"id":91102367,"identity":"65a21863-dc10-4a9d-bba8-0d526285ca9c","added_by":"auto","created_at":"2025-09-11 14:57:37","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":8834,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7473073/v1/84bc6bd9cb12778969d29601.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Hierarchical Clustering-Based Coarse-to-Fine Classification Framework for Microbial Protein Function Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"bmc-bioinformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binf","sideBox":"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/binf","title":"BMC Bioinformatics","twitterHandle":"@BMC_Bioinformatics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Protein Function Prediction, Deep learning, Gene Ontology, Coarse-to-Fine Classification, Protein Sequence, Long-tail Label Prediction","lastPublishedDoi":"10.21203/rs.3.rs-7473073/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7473073/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e\u003cp\u003eAccurate prediction of microbial protein functions is essential for understanding microbial physiology, discovering novel probiotics, and driving biotechnological innovation. However, protein function prediction remains challenging due to the hierarchical and class-imbalanced nature of functional labels, particularly in large-scale annotations such as Enzyme Commission (EC) numbers and Gene Ontology (GO) terms. Most existing deep learning approaches fail to adequately address the long-tail distribution problem.\u003c/p\u003e\u003ch2\u003eMethods:\u003c/h2\u003e\u003cp\u003eWe propose a Hierarchical Cascaded Context Network (HCCN) that explicitly models functional hierarchies and emphasizes prediction of low-frequency (long-tail) labels. For EC classification, we design a coarse-to-fine network that captures parent\u0026ndash;child dependencies among hierarchical labels. For GO prediction, we construct a semantically grounded hierarchical structure using ontology embedding and clustering, and develop an attention-based multi-level cascade predictor to exploit structured dependencies across Biological Process (BPO), Molecular Function (MFO), and Cellular Component (CCO). To mitigate label imbalance, we introduce a dynamic resampling strategy and a hierarchical loss weighting mechanism, which enforce inter-level regularization and enhance sensitivity to rare functions.\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e\u003cp\u003eExperimental results show that HCCN consistently outperforms traditional sequence-alignment methods (e.g., DIAMOND, BLAST) and baseline neural networks (MLP) across all major functional categories. On the full test set, HCCN achieves AUPR gains of up to 9.3% (EC), 10.6% (BPO), 6.9% (MFO), and 5.4% (CCO) over the best baseline. For low-frequency labels, HCCN demonstrates strong few-shot generalization, with improvements of +\u0026thinsp;11.2% (EC low) and +\u0026thinsp;10.9% (BPO low) in mAUPR.\u003c/p\u003e\u003ch2\u003eConclusions:\u003c/h2\u003e\u003cp\u003eThe proposed HCCN framework provides an effective solution to hierarchical and imbalanced protein function prediction, significantly improving performance on long-tail functional labels. Code and data are publicly available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/YangLab-BUPT/HCCN\u003c/span\u003e\u003cspan address=\"https://github.com/YangLab-BUPT/HCCN\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e","manuscriptTitle":"Hierarchical Clustering-Based Coarse-to-Fine Classification Framework for Microbial Protein Function Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-11 14:57:32","doi":"10.21203/rs.3.rs-7473073/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-09T13:13:20+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-04T22:35:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"196366964813373134643969699004808456029","date":"2025-09-24T02:22:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"32227526411825339307326680584974504911","date":"2025-09-21T23:14:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"65276440693225446148366942049531463632","date":"2025-09-21T02:11:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-06T01:31:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135702925704310671052440471103229091955","date":"2025-09-05T00:23:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-04T14:55:01+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-04T10:19:04+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-28T02:01:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-28T02:00:50+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Bioinformatics","date":"2025-08-27T15:19:20+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-bioinformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binf","sideBox":"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/binf","title":"BMC Bioinformatics","twitterHandle":"@BMC_Bioinformatics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a3a76671-1bac-4f04-80e2-cf9dae3d808b","owner":[],"postedDate":"September 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-22T16:01:21+00:00","versionOfRecord":{"articleIdentity":"rs-7473073","link":"https://doi.org/10.1186/s12859-025-06326-7","journal":{"identity":"bmc-bioinformatics","isVorOnly":false,"title":"BMC Bioinformatics"},"publishedOn":"2025-12-20 15:57:44","publishedOnDateReadable":"December 20th, 2025"},"versionCreatedAt":"2025-09-11 14:57:32","video":"","vorDoi":"10.1186/s12859-025-06326-7","vorDoiUrl":"https://doi.org/10.1186/s12859-025-06326-7","workflowStages":[]},"version":"v1","identity":"rs-7473073","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7473073","identity":"rs-7473073","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}