Feature-Optimized Machine Learning Benchmarking for Protein Interface Prediction in Permanent Homodimer Complexes with Distinct Structural Features

Feature-Optimized Machine Learning Benchmarking for Protein Interface Prediction in Permanent Homodimer Complexes with Distinct Structural Features | 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 Feature-Optimized Machine Learning Benchmarking for Protein Interface Prediction in Permanent Homodimer Complexes with Distinct Structural Features Tayyip Topuz, Zeki Erdem, Halil Bisgin, E. Demet Akten This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8559148/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 Accurate prediction of protein-protein interaction interfaces is critical for understanding molecular recognition and guiding therapeutic design. This study presents a comprehensive machine learning pipeline for predicting interface residues in permanent homodimeric protein complexes. Using a curated dataset of 1,311 homodimers, we benchmarked six widely used machine learning algorithms and identified Multilayer Perceptron and XGBoost as top performers, achieving Matthews Correlation Coefficients (MCC) exceeding 0.93. To enhance interpretability and efficiency, we employed recursive feature elimination and derived a minimal set of six biologically meaningful features, including solvent accessibility, surface roughness, planarity, and average protrusion index, that retained high predictive power (MCC > 0.90). Structurally stratified models tailored to α-helical, β-strand, and membrane proteins demonstrated comparable or improved accuracy relative to generalized models, particularly when utilizing the reduced feature subset. We further validated our approach on an external heterodimer complex (PDB ID: 9ETL), where structurally specialized models generalized well, confirming robustness beyond the training domain. The results highlight the importance of structural context in interface prediction and demonstrate that compact, structure-aware models can achieve high accuracy while reducing computational complexity. This work provides a scalable, interpretable, and biologically informed approach to protein interface prediction, with implications for large-scale structural descriptor, drug target characterization, and protein engineering applications. Biological sciences/Computational biology and bioinformatics Biological sciences/Drug discovery Physical sciences/Mathematics and computing Protein-Protein Interface (PPI) Secondary Structure Type Membrane Protein Structure Prediction Machine Learning Feature Engineering Hyperparameter Optimization Full Text Additional Declarations No competing interests reported. Supplementary Files supplementarymaterials.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 24 Mar, 2026 Reviews received at journal 13 Mar, 2026 Reviews received at journal 12 Mar, 2026 Reviewers agreed at journal 19 Feb, 2026 Reviewers agreed at journal 19 Feb, 2026 Reviewers invited by journal 30 Jan, 2026 Editor assigned by journal 10 Jan, 2026 Submission checks completed at journal 10 Jan, 2026 First submitted to journal 09 Jan, 2026 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-8559148","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":583757116,"identity":"0253c949-548b-487b-8929-55a037e93507","order_by":0,"name":"Tayyip Topuz","email":"","orcid":"","institution":"Kadir Has University","correspondingAuthor":false,"prefix":"","firstName":"Tayyip","middleName":"","lastName":"Topuz","suffix":""},{"id":583757117,"identity":"983d3fe7-4e99-4c07-8866-05172ffb172a","order_by":1,"name":"Zeki Erdem","email":"","orcid":"","institution":"Kadir Has University","correspondingAuthor":false,"prefix":"","firstName":"Zeki","middleName":"","lastName":"Erdem","suffix":""},{"id":583757118,"identity":"f0b3b708-54c0-475b-8840-ed2dae027f93","order_by":2,"name":"Halil Bisgin","email":"","orcid":"","institution":"University of Michigan- Flint","correspondingAuthor":false,"prefix":"","firstName":"Halil","middleName":"","lastName":"Bisgin","suffix":""},{"id":583757119,"identity":"e13af66d-744a-4569-820d-01a14b47fda3","order_by":3,"name":"E. Demet Akten","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYDCCAxDKAIgZHzY2wDkEtACxAQ8DA7MhQksCcVrYJInSwne8O/HxBwY7Y3v23mOVM3fY2TOwN2+TYPxxD6cWyTNnNxscYEg24+E5l3Zz45nkxAaeY2USDAnFOLUY3MjdJnGAgdmGRyLH7ObDNuYEBiADqAW3ywzuv93+4wBDvQ2P/Buzwodt9fYMQAZ+LTd4twG9f9iMR4LHjHFj22HGBiADrxbJM7mbJc4YHDfmOZNjLDmz7XhiG09asUVCGm4tfMfPbvxQUVFt2N5+xvBjb1u1PT/74Y03Ptjg1gJ1HhKbDUQQ0jAKRsEoGAWjAD8AANjWU+I81en/AAAAAElFTkSuQmCC","orcid":"","institution":"Kadir Has University","correspondingAuthor":true,"prefix":"","firstName":"E.","middleName":"Demet","lastName":"Akten","suffix":""}],"badges":[],"createdAt":"2026-01-09 09:23:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8559148/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8559148/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101752913,"identity":"27a894b7-1588-43cd-bc05-66a91a459ed3","added_by":"auto","created_at":"2026-02-03 10:38:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1346236,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptTopuzetalJan2026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8559148/v1_covered_5a08c72e-4a0f-48e2-9166-e01269098c88.pdf"},{"id":101625626,"identity":"c72be635-f24f-427f-8b63-1727968b1c74","added_by":"auto","created_at":"2026-02-02 03:50:39","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":900553,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-8559148/v1/4b4fc1dea9f965ce1640a72c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Feature-Optimized Machine Learning Benchmarking for Protein Interface Prediction in Permanent Homodimer Complexes with Distinct Structural Features","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Protein-Protein Interface (PPI), Secondary Structure Type, Membrane, Protein Structure Prediction, Machine Learning, Feature Engineering, Hyperparameter Optimization","lastPublishedDoi":"10.21203/rs.3.rs-8559148/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8559148/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccurate prediction of protein-protein interaction interfaces is critical for understanding molecular recognition and guiding therapeutic design. This study presents a comprehensive machine learning pipeline for predicting interface residues in permanent homodimeric protein complexes. Using a curated dataset of 1,311 homodimers, we benchmarked six widely used machine learning algorithms and identified Multilayer Perceptron and XGBoost as top performers, achieving Matthews Correlation Coefficients (MCC) exceeding 0.93. To enhance interpretability and efficiency, we employed recursive feature elimination and derived a minimal set of six biologically meaningful features, including solvent accessibility, surface roughness, planarity, and average protrusion index, that retained high predictive power (MCC\u0026thinsp;\u0026gt;\u0026thinsp;0.90). Structurally stratified models tailored to α-helical, β-strand, and membrane proteins demonstrated comparable or improved accuracy relative to generalized models, particularly when utilizing the reduced feature subset. We further validated our approach on an external heterodimer complex (PDB ID: 9ETL), where structurally specialized models generalized well, confirming robustness beyond the training domain. The results highlight the importance of structural context in interface prediction and demonstrate that compact, structure-aware models can achieve high accuracy while reducing computational complexity. This work provides a scalable, interpretable, and biologically informed approach to protein interface prediction, with implications for large-scale structural descriptor, drug target characterization, and protein engineering applications.\u003c/p\u003e","manuscriptTitle":"Feature-Optimized Machine Learning Benchmarking for Protein Interface Prediction in Permanent Homodimer Complexes with Distinct Structural Features","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-02 03:50:34","doi":"10.21203/rs.3.rs-8559148/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-24T12:01:24+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-14T02:52:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-12T08:36:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"132812386060539316781428431758675872634","date":"2026-02-20T00:41:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328568329955881964535207896485853829109","date":"2026-02-19T11:23:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-30T06:48:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-10T05:09:37+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-10T05:07:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-09T09:06:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8b5b3686-e078-4ecd-808d-48ae0ed105ad","owner":[],"postedDate":"February 2nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62104738,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":62104739,"name":"Biological sciences/Drug discovery"},{"id":62104740,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-17T14:08:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-02 03:50:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8559148","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8559148","identity":"rs-8559148","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}