Unveiling Molecular Moieties through Hierarchical Graph Explainability

Unveiling Molecular Moieties through Hierarchical Graph Explainability | 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 Unveiling Molecular Moieties through Hierarchical Graph Explainability Paolo Sortino, Salvatore Contino, Ugo Perricone, Roberto Pirrone This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4206999/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 Background : Graph Neural Networks (GNN) have emerged in very recent years as a powerful tool for supporting in silico Virtual Screening. In this work we present a GNN which uses Graph Convolutional (GC) architectures to achieve very accurate multi-target screening. As novelty compared to state-of-the-art we devised a hierarchical Explainable Artificial Intelligence (XAI) technique to catch information directly at atom, ring, and whole molecule level by leveraging the message passing mechanism. In this way, we find the most relevant moieties involved in bioactivity prediction. Results : We report a state-of-the-art GNN classifier on twenty Cyclin-dependent Kinase targets in support of VS. Our classifier outperforms previous SOTA approaches proposed by the authors. Moreover, a CDK1-only high-sensitivity version of the GNN has been designed to use our explainer in order to avoid the inherent bias of multi-class models. The hierarchical explainer has been validated by comparison with literature, experimental and docking data on 19 approved drugs on CDK1. Our explainer provided information in accordance to the docking analysis for 17 out of the 19 test drugs. Conclusion : Our approach is a valid support for shortening both the screening and the hit-to-lead phase. Detailed knowledge about the molecular substructures that play a role in the inhibitory action, can help the computational chemist to gain insights into the pharmacophoric function of the molecule also for repurposing tasks. Graph Neural Networks Graph Explainability Computer-Aided Drug Design (CADD) Virtual Screening Molecular moieties Molecular Docking Full Text Additional Declarations No competing interests reported. Supplementary Files supplementarymaterial.pdf 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-4206999","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":287301915,"identity":"b696f4a5-ab82-4556-96fb-042d9e6aa9c5","order_by":0,"name":"Paolo Sortino","email":"","orcid":"","institution":"University of Palermo","correspondingAuthor":false,"prefix":"","firstName":"Paolo","middleName":"","lastName":"Sortino","suffix":""},{"id":287301916,"identity":"7d31abe4-a8f4-45d2-80cf-aaea7e6e869c","order_by":1,"name":"Salvatore Contino","email":"data:image/png;base64,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","orcid":"","institution":"University of Palermo","correspondingAuthor":true,"prefix":"","firstName":"Salvatore","middleName":"","lastName":"Contino","suffix":""},{"id":287301917,"identity":"9311f945-04b5-4ca5-b967-1f6cf05e3813","order_by":2,"name":"Ugo Perricone","email":"","orcid":"","institution":"Fondazione Ri.MED","correspondingAuthor":false,"prefix":"","firstName":"Ugo","middleName":"","lastName":"Perricone","suffix":""},{"id":287301918,"identity":"1303ad96-ef02-4117-8237-33a0a0f4ffa1","order_by":3,"name":"Roberto Pirrone","email":"","orcid":"","institution":"University of Palermo","correspondingAuthor":false,"prefix":"","firstName":"Roberto","middleName":"","lastName":"Pirrone","suffix":""}],"badges":[],"createdAt":"2024-04-02 13:20:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4206999/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4206999/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":65030571,"identity":"0506266a-3b98-4bda-93d2-5aa24b091102","added_by":"auto","created_at":"2024-09-22 21:31:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":878632,"visible":true,"origin":"","legend":"","description":"","filename":"HGE.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4206999/v1_covered_d927e801-4f54-48ff-9b6e-58b8df2c31b3.pdf"},{"id":54260106,"identity":"0cfacac9-4cfa-4f90-ba67-bec44c767a8b","added_by":"auto","created_at":"2024-04-08 02:57:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1459023,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4206999/v1/8e080f9783e654b144980ff1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Unveiling Molecular Moieties through Hierarchical Graph Explainability","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"Graph Neural Networks, Graph Explainability, Computer-Aided Drug Design (CADD), Virtual Screening, Molecular moieties, Molecular Docking","lastPublishedDoi":"10.21203/rs.3.rs-4206999/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4206999/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Graph Neural Networks (GNN) have emerged in very recent years as a powerful tool for supporting in silico Virtual Screening. In this work we present a GNN which uses Graph Convolutional (GC) architectures to achieve very accurate multi-target screening. As novelty compared to state-of-the-art we devised a hierarchical Explainable Artificial Intelligence (XAI) technique to catch information directly at atom, ring, and whole molecule level by leveraging the message passing mechanism. In this way, we find the most relevant moieties involved in bioactivity prediction.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: We report a state-of-the-art GNN classifier on twenty Cyclin-dependent Kinase targets in support of VS. Our classifier outperforms previous SOTA approaches proposed by the authors. Moreover, a CDK1-only high-sensitivity version of the GNN has been designed to use our explainer in order to avoid the inherent bias of multi-class models. The hierarchical explainer has been validated by comparison with literature, experimental and docking data on 19 approved drugs on CDK1. Our explainer provided information in accordance to the docking analysis for 17 out of the 19 test drugs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: Our approach is a valid support for shortening both the screening and the hit-to-lead phase. Detailed knowledge about the molecular substructures that play a role in the inhibitory action, can help the computational chemist to gain insights into the pharmacophoric function of the molecule also for repurposing tasks.\u003c/p\u003e","manuscriptTitle":"Unveiling Molecular Moieties through Hierarchical Graph Explainability","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-08 02:57:35","doi":"10.21203/rs.3.rs-4206999/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":"bec7c632-2fd4-448e-8338-b81e436fb3d9","owner":[],"postedDate":"April 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-09-22T21:23:21+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-08 02:57:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4206999","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4206999","identity":"rs-4206999","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}