Extrapolation of Machine-Learning Interatomic Potentials for Organic and Polymeric Systems

Extrapolation of Machine-Learning Interatomic Potentials for Organic and Polymeric Systems | 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 Extrapolation of Machine-Learning Interatomic Potentials for Organic and Polymeric Systems Natalie Hooven, Arthur Lin, Rose Cersonsky This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7744194/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 Machine-Learning Interatomic Potentials (MLIPs) have surged in popularity due to their promise of expanding the spatiotemporal scales possible for simulating molecules with high fidelity. The accuracy of any MLIP is dependent on the data used for its training; thus, for large molecules, like polymers, where accurate training data is prohibitively difficult to obtain, it becomes necessary to pursue non-traditional methods to construct MLIPs, many of which are based on constructing MLIPs using smaller, analogous chemical systems. However, we have yet to understand the limits to which smaller molecules can be used as a proxy for extrapolating macromolecular energetics. Here, we provide a ''control study'' for such experiments, exploring the ability of MLIP approaches to extrapolate between n=1-8 n-polyalkanes at identical conditions. Through Principal Covariates Classification, we quantitatively demonstrate how convergence in chemical environments between training and testing datasets coincides with an MLIP's transferability. Additionally, we show how careful attention to the construction of an MLIP's neighbor list can promote greater transferability when considering various levels of the energetic hierarchy. Our results establish a roadmap for how one can create transferable MLIPs for macromolecular systems without the prohibitive cost of constructing system-specific training data. Physical sciences/Chemistry Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. 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-7744194","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":534313469,"identity":"732d67fe-32f6-48bb-ac35-1f8284ea6643","order_by":0,"name":"Natalie Hooven","email":"","orcid":"","institution":"University of Wisconsin–Madison","correspondingAuthor":false,"prefix":"","firstName":"Natalie","middleName":"","lastName":"Hooven","suffix":""},{"id":534313470,"identity":"2f3ddfeb-f192-465f-b921-7e5d2ae450d7","order_by":1,"name":"Arthur Lin","email":"","orcid":"","institution":"University of Wisconsin–Madison","correspondingAuthor":false,"prefix":"","firstName":"Arthur","middleName":"","lastName":"Lin","suffix":""},{"id":534313471,"identity":"c53e968d-5c65-495d-84d3-d0f26c13948b","order_by":2,"name":"Rose Cersonsky","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYLADxgYGGwY2MJONeC1ppGs5DGXi0WLe3v7wAUONTb58e+/hjzPbzif28R9+wPCh7DBOLTJnzhgbMBxLs9xw5lya5Ma224ltEmkGjDPO4dYiIZHDJsHYcNjAQCLHjPFh2+3cNgkeBmbeNnxa0p//YGz4byA/I8f448O2c7lt/GcYmP/i1ZJgBvT2AQOGGzkGQIcdyG1jyGFgZsSnheeMsUTCsWQDgzNnzCRnnEuuB/nlYM+5dNxa2NsffvhQY2cg395j/LGnzM5Yvv/wwwc/yqxxagGDBBiDERodB/CrRwF/SFA7CkbBKBgFIwYAAGF3U/UGEwt/AAAAAElFTkSuQmCC","orcid":"","institution":"University of Wisconsin–Madison","correspondingAuthor":true,"prefix":"","firstName":"Rose","middleName":"","lastName":"Cersonsky","suffix":""}],"badges":[],"createdAt":"2025-09-29 16:53:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7744194/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7744194/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94456258,"identity":"c49d85c9-a1e8-4688-9be4-a4e077805b03","added_by":"auto","created_at":"2025-10-27 14:44:22","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4955,"visible":true,"origin":"","legend":"","description":"","filename":"e6e4d844568148448e33caf0a963a7fc.json","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/9676ca01a3b2d08d65ce9553.json"},{"id":94456680,"identity":"ff420904-8de5-401a-a1c3-ac4b683513dd","added_by":"auto","created_at":"2025-10-27 14:44:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2715073,"visible":true,"origin":"","legend":"","description":"","filename":"ExtrapolationofAlkanePotentials1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/1e03fac2d29f030cb229de35.pdf"},{"id":94456746,"identity":"cbc0cec5-1b6f-4dc7-8fd3-fc78f90ffdea","added_by":"auto","created_at":"2025-10-27 14:45:05","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119873,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1a.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/0f269c899c96643db62faa35.png"},{"id":94455749,"identity":"622bec9c-e056-42b7-bc0b-7f3918b6e495","added_by":"auto","created_at":"2025-10-27 14:43:54","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83769,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1b.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/a78be20a50a16544191027e6.png"},{"id":94456572,"identity":"366139fd-9418-4dde-a6ef-f8347041a259","added_by":"auto","created_at":"2025-10-27 14:44:42","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66212,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1c.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/bcb6ca67a1d4d25f04bab1a9.png"},{"id":94456255,"identity":"c2128962-af6a-4aed-9c2d-1c08eb6d83b2","added_by":"auto","created_at":"2025-10-27 14:44:22","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":126517,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1d.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/2db572f9d55fd89b7a4b46d4.png"},{"id":94456524,"identity":"a319b500-99a0-402c-a189-d4fd0e83715e","added_by":"auto","created_at":"2025-10-27 14:44:39","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":245407,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/c77f6a88d1f9dd4d5b640bac.pdf"},{"id":94456295,"identity":"94bd137d-d513-4c08-8df0-36a4cc4897bb","added_by":"auto","created_at":"2025-10-27 14:44:23","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117466,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3a.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/ffa3cc471e3f2f2222953047.png"},{"id":94455746,"identity":"9490976c-ff19-4005-8881-efeb7de3f1cb","added_by":"auto","created_at":"2025-10-27 14:43:54","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109474,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3b.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/6dfe2c02a858c7554f40ee9a.png"},{"id":94489430,"identity":"d07afec0-656a-4ade-8cea-5237c1b52aa9","added_by":"auto","created_at":"2025-10-27 17:04:39","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103790,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1a.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/6913114cb4cedcaf753ec5b5.png"},{"id":94456589,"identity":"40ab9e25-7371-48f9-8af1-777ab4032d9e","added_by":"auto","created_at":"2025-10-27 14:44:43","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71016,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1b.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/48344d39474dd7d3649f8659.png"},{"id":94455451,"identity":"ed7f4f40-c639-4ff6-be1f-100919f92a25","added_by":"auto","created_at":"2025-10-27 14:43:44","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":73524,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/131ecb906577b29768c7774f.png"},{"id":94455891,"identity":"a522ddef-be9f-49ee-86b2-442e0dfbfb81","added_by":"auto","created_at":"2025-10-27 14:44:01","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117177,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3a.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/872ac4fab685757165836101.png"},{"id":94456590,"identity":"4569a291-aa10-4609-8e05-701557f8ff59","added_by":"auto","created_at":"2025-10-27 14:44:43","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90100,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3b.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/694fae6a8999617489fcacbf.png"},{"id":94456624,"identity":"a0d1d212-952d-4a0c-a909-dd8e680837ba","added_by":"auto","created_at":"2025-10-27 14:44:51","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71325,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3c.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/6c2a13f1911a4bc3a6790900.png"},{"id":94456749,"identity":"655725ad-e273-4097-bb4e-d74197bd9fa1","added_by":"auto","created_at":"2025-10-27 14:45:06","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127253,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3d.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/d3df138d8969553b7668ac0d.png"},{"id":94455876,"identity":"0ef178cc-708b-4335-a0e4-dd1056f10fe5","added_by":"auto","created_at":"2025-10-27 14:44:00","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":391386,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4a.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/79a8c2caff1bd612833845ee.pdf"},{"id":94456164,"identity":"0eff603f-c85a-4dff-8b57-487f8c6cb5f9","added_by":"auto","created_at":"2025-10-27 14:44:18","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":198121,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4b.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/1c4798eef23b4a0f85e00fab.pdf"},{"id":94456688,"identity":"faf7dc42-865a-4a25-a74b-ec83fe693409","added_by":"auto","created_at":"2025-10-27 14:44:59","extension":"pdf","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":210789,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4c.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/cf49b6a267de3267afc2b419.pdf"},{"id":94456522,"identity":"daff1fed-9139-4edc-a10f-20eea5fb2150","added_by":"auto","created_at":"2025-10-27 14:44:38","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108892,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS5a.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/f2d70f5f6eee69ef116aac53.png"},{"id":94456252,"identity":"6ffaffdc-5039-48da-8186-c6e749c96ad8","added_by":"auto","created_at":"2025-10-27 14:44:21","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112893,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS5b.png","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/e9cf9c84c568fd65e52eaa13.png"},{"id":94456324,"identity":"41626bdc-785a-4e5f-985f-529c55ecf917","added_by":"auto","created_at":"2025-10-27 14:44:27","extension":"pdf","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268472,"visible":true,"origin":"","legend":"","description":"","filename":"PolymericExtrapolationCoverLetter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/c0fe5dfd7861c43d2a9efcb5.pdf"},{"id":94456600,"identity":"b91403c9-ce17-4523-8ab3-02fc2cdea7d0","added_by":"auto","created_at":"2025-10-27 14:44:46","extension":"pdf","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104455,"visible":true,"origin":"","legend":"","description":"","filename":"TOC.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/f0474c00576f02b5526c879c.pdf"},{"id":94456758,"identity":"f7d40f09-6eb2-484c-a21e-8ec490473c7c","added_by":"auto","created_at":"2025-10-27 14:45:06","extension":"xml","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76876,"visible":true,"origin":"","legend":"","description":"","filename":"e6e4d844568148448e33caf0a963a7fc1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1/a47e17d43da8da96b4ab51ee.xml"},{"id":94599940,"identity":"307d8610-cbac-47f3-8061-d7924ad70487","added_by":"auto","created_at":"2025-10-28 19:08:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2098934,"visible":true,"origin":"","legend":"","description":"","filename":"ExtrapolationofAlkanePotentials1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7744194/v1_covered_83b2d2e3-b134-4e00-b897-2f862c615453.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Extrapolation of Machine-Learning Interatomic Potentials for Organic and Polymeric Systems","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":"","lastPublishedDoi":"10.21203/rs.3.rs-7744194/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7744194/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Machine-Learning Interatomic Potentials (MLIPs) have surged in popularity due to their promise of expanding the spatiotemporal scales possible for simulating molecules with high fidelity. The accuracy of any MLIP is dependent on the data used for its training; thus, for large molecules, like polymers, where accurate training data is prohibitively difficult to obtain, it becomes necessary to pursue non-traditional methods to construct MLIPs, many of which are based on constructing MLIPs using smaller, analogous chemical systems. However, we have yet to understand the limits to which smaller molecules can be used as a proxy for extrapolating macromolecular energetics. Here, we provide a ''control study'' for such experiments, exploring the ability of MLIP approaches to extrapolate between n=1-8 n-polyalkanes at identical conditions. Through Principal Covariates Classification, we quantitatively demonstrate how convergence in chemical environments between training and testing datasets coincides with an MLIP's transferability. Additionally, we show how careful attention to the construction of an MLIP's neighbor list can promote greater transferability when considering various levels of the energetic hierarchy. Our results establish a roadmap for how one can create transferable MLIPs for macromolecular systems without the prohibitive cost of constructing system-specific training data.","manuscriptTitle":"Extrapolation of Machine-Learning Interatomic Potentials for Organic and Polymeric Systems","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-27 11:39:59","doi":"10.21203/rs.3.rs-7744194/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":"0b3bc7fa-7c1e-4eed-896e-6ad1e3844137","owner":[],"postedDate":"October 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56811522,"name":"Physical sciences/Chemistry"},{"id":56811523,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56811524,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-10-28T18:56:50+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-27 11:39:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7744194","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7744194","identity":"rs-7744194","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}