Site-Conditioned Edit Chemistry for Cyclic Peptide Permeability Modeling

Site-Conditioned Edit Chemistry for Cyclic Peptide Permeability Modeling | 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 Site-Conditioned Edit Chemistry for Cyclic Peptide Permeability Modeling Minrui Chen, Guanjie Zou, Minqi Chen, Bowen Song, Hiroto Saigo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9541490/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Localized chemical edits such as noncanonical substitutions, post-translational modifications, cyclization, and linker-mediated conjugation can strongly alter cyclic peptide permeability. However, most predictive studies do not explicitly separate the contribution of edit chemistry itself from the contribution of assigning that chemistry to a specific peptide site. Here, we study this question in a deliberately restricted curated known-site regime, in which modification sites can be resolved after rule-based preprocessing, and formulate the resulting task as chemistry--site contribution analysis under explicit site observability. Using the assay-harmonized CycPeptMPDB-PAMPA dataset, we compare sequence-only, chemistry-only, and site-aware descriptor models through controlled feature-block comparisons. We further interrogate the learned signal using wrong-site and coarse-site controls, graded site perturbation, feature ablation, generalized additive modeling, anchor-sensitive kernel regression, parser auditing, and a harder sequence-cluster split.Across the standard random split, chemistry-aware descriptor modeling substantially outperforms sequence-only prediction, indicating that edit chemistry provides the dominant predictive signal for passive permeability in this dataset. Adding site-aware features yields a reproducible but quantitatively modest refinement, improving test \((R^2)\) from 0.433 to 0.472. Perturbation controls show that this refinement is strongly position-sensitive: wrong-site, coarse-site, and graded site-shift settings all degrade performance while holding edit chemistry fixed. Feature ablation further indicates that the strongest individually effective site-aware component in the current descriptor family is site-position statistics, whereas anchor-site residue-composition and residue-property summaries contribute mainly in combination with positional information. However, under a harder sequence-cluster split, the site-aware gain nearly vanishes, with test \((R^2)\) changing only from 0.268 to 0.270. This result indicates that the current site-conditioned benefit is context-dependent and has limited transfer across stronger scaffold shift.We therefore position the present study not as a universal peptide-property predictor and not as a solution to uncertain-site inference, but as a mechanistic analysis of chemistry-versus-site contributions in a curated known-site cyclic peptide permeability setting. The main contribution is to quantify the present scale, locality, and boundary conditions of explicit site conditioning within an interpretable descriptor framework. The results support a chemistry-dominant view of permeability prediction, in which site information provides a local within-regime refinement rather than a broadly transferable predictive advantage under scaffold shift. cyclic peptides permeability prediction edit chemistry site-conditioned modeling chemical descriptors cheminformatics mechanistic modeling Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 13 May, 2026 Reviewers agreed at journal 12 May, 2026 Reviewers agreed at journal 11 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers invited by journal 07 May, 2026 Editor assigned by journal 29 Apr, 2026 Submission checks completed at journal 29 Apr, 2026 First submitted to journal 27 Apr, 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-9541490","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":641551328,"identity":"15ac1548-b38d-49b6-be32-75ceef2b40d6","order_by":0,"name":"Minrui Chen","email":"","orcid":"","institution":"Kyushu University","correspondingAuthor":false,"prefix":"","firstName":"Minrui","middleName":"","lastName":"Chen","suffix":""},{"id":641551332,"identity":"804810e9-504e-449e-bd45-2e57483664a0","order_by":1,"name":"Guanjie Zou","email":"","orcid":"","institution":"The University of Tokyo","correspondingAuthor":false,"prefix":"","firstName":"Guanjie","middleName":"","lastName":"Zou","suffix":""},{"id":641551334,"identity":"9e026c1e-03bd-4e0d-b665-3c71d314bd7e","order_by":2,"name":"Minqi Chen","email":"","orcid":"","institution":"West China Hospital, Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Minqi","middleName":"","lastName":"Chen","suffix":""},{"id":641551335,"identity":"22ee00a6-3f0a-4154-81cc-7727a758fef3","order_by":3,"name":"Bowen Song","email":"","orcid":"","institution":"Kyushu University","correspondingAuthor":false,"prefix":"","firstName":"Bowen","middleName":"","lastName":"Song","suffix":""},{"id":641551340,"identity":"5a542b7c-8cf7-47e4-8af8-63c1064c0725","order_by":4,"name":"Hiroto Saigo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIie3Rv0rDQBzA8V+IXJaL80kofYVIF/M2+VHQrUuXDqUcFH5dAlkV1Mfo2sJBpj7ADS6hIA4OcSs0g3cpLpWLjoL3XXIX+HD/AHy+P1kgAWYCIsnsbPv1m/1AdgL49vfEKoJvxN1wpZZvH883Ew7sNuHtC0qRB80BLicuku6Qsoe1mBpSJTG9WhJeFcCmTgJIo3gtcAMRJbFUuNE5JOYsKF0bK2tDHgUWlvBWdasc+whoXO5jaYnZGGcdYb2rpLqm4L4yJGTj7InUSPKasiJ1n2VY3u2bZr7AIqJr/d6qgYzGSh9mlfPGTEx0nxAuToPucdMq7yFhcz6wzfuIz+fz/a8+AUuGVS9TY+qxAAAAAElFTkSuQmCC","orcid":"","institution":"Kyushu University","correspondingAuthor":true,"prefix":"","firstName":"Hiroto","middleName":"","lastName":"Saigo","suffix":""}],"badges":[],"createdAt":"2026-04-27 12:09:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9541490/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9541490/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109760163,"identity":"34f0ba57-d6ac-4732-9e16-6ce8fe0118ab","added_by":"auto","created_at":"2026-05-22 07:28:14","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":539908,"visible":true,"origin":"","legend":"","description":"","filename":"SiteConditionedEditChemistryforCyclicPeptidePermeabilityModeling.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9541490/v1_covered_ae0dcb6a-633f-4bc9-928f-51f0f7c33257.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Site-Conditioned Edit Chemistry for Cyclic Peptide Permeability Modeling","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":"journal-of-cheminformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"chin","sideBox":"Learn more about [Journal of Cheminformatics](https://jcheminf.biomedcentral.com/)","snPcode":"13321","submissionUrl":"https://submission.nature.com/new-submission/13321/3","title":"Journal of Cheminformatics","twitterHandle":"@jcheminf","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"cyclic peptides, permeability prediction, edit chemistry, site-conditioned modeling, chemical descriptors, cheminformatics, mechanistic modeling","lastPublishedDoi":"10.21203/rs.3.rs-9541490/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9541490/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLocalized chemical edits such as noncanonical substitutions, post-translational modifications, cyclization, and linker-mediated conjugation can strongly alter cyclic peptide permeability. However, most predictive studies do not explicitly separate the contribution of edit chemistry itself from the contribution of assigning that chemistry to a specific peptide site. Here, we study this question in a deliberately restricted curated known-site regime, in which modification sites can be resolved after rule-based preprocessing, and formulate the resulting task as chemistry--site contribution analysis under explicit site observability. Using the assay-harmonized CycPeptMPDB-PAMPA dataset, we compare sequence-only, chemistry-only, and site-aware descriptor models through controlled feature-block comparisons. We further interrogate the learned signal using wrong-site and coarse-site controls, graded site perturbation, feature ablation, generalized additive modeling, anchor-sensitive kernel regression, parser auditing, and a harder sequence-cluster split.Across the standard random split, chemistry-aware descriptor modeling substantially outperforms sequence-only prediction, indicating that edit chemistry provides the dominant predictive signal for passive permeability in this dataset. Adding site-aware features yields a reproducible but quantitatively modest refinement, improving test \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\((R^2)\\)\u003c/span\u003e\u003c/span\u003e from 0.433 to 0.472. Perturbation controls show that this refinement is strongly position-sensitive: wrong-site, coarse-site, and graded site-shift settings all degrade performance while holding edit chemistry fixed. Feature ablation further indicates that the strongest individually effective site-aware component in the current descriptor family is site-position statistics, whereas anchor-site residue-composition and residue-property summaries contribute mainly in combination with positional information. However, under a harder sequence-cluster split, the site-aware gain nearly vanishes, with test \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\((R^2)\\)\u003c/span\u003e\u003c/span\u003e changing only from 0.268 to 0.270. This result indicates that the current site-conditioned benefit is context-dependent and has limited transfer across stronger scaffold shift.We therefore position the present study not as a universal peptide-property predictor and not as a solution to uncertain-site inference, but as a mechanistic analysis of chemistry-versus-site contributions in a curated known-site cyclic peptide permeability setting. The main contribution is to quantify the present scale, locality, and boundary conditions of explicit site conditioning within an interpretable descriptor framework. The results support a chemistry-dominant view of permeability prediction, in which site information provides a local within-regime refinement rather than a broadly transferable predictive advantage under scaffold shift.\u003c/p\u003e","manuscriptTitle":"Site-Conditioned Edit Chemistry for Cyclic Peptide Permeability Modeling","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-18 07:00:43","doi":"10.21203/rs.3.rs-9541490/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T09:25:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294158526574194356646731519186430140947","date":"2026-05-12T11:56:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56293532719599147732992679556516531935","date":"2026-05-11T07:37:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"198297120184190625875027076896724901109","date":"2026-05-07T14:55:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-07T14:29:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-29T04:46:32+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-29T04:45:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Cheminformatics","date":"2026-04-27T11:57:08+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-cheminformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"chin","sideBox":"Learn more about [Journal of Cheminformatics](https://jcheminf.biomedcentral.com/)","snPcode":"13321","submissionUrl":"https://submission.nature.com/new-submission/13321/3","title":"Journal of Cheminformatics","twitterHandle":"@jcheminf","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"758b0e9b-3724-4b5f-a049-a4c9cb619c6e","owner":[],"postedDate":"May 18th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-13T09:25:02+00:00","index":21,"fulltext":""},{"type":"reviewerAgreed","content":"294158526574194356646731519186430140947","date":"2026-05-12T11:56:35+00:00","index":20,"fulltext":""},{"type":"reviewerAgreed","content":"56293532719599147732992679556516531935","date":"2026-05-11T07:37:44+00:00","index":17,"fulltext":""},{"type":"reviewerAgreed","content":"198297120184190625875027076896724901109","date":"2026-05-07T14:55:51+00:00","index":14,"fulltext":""},{"type":"reviewersInvited","content":"7","date":"2026-05-07T14:29:01+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-18T07:00:43+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-18 07:00:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9541490","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9541490","identity":"rs-9541490","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}