{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T02:45:00Z","timestamp":1785897900859,"version":"3.56.0"},"posted":{"date-parts":[[2026,8,1]]},"group-title":"Biophysics","reference-count":49,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000057","name":"National Institute of General Medical Sciences","doi-asserted-by":"crossref","award":["R35GM128906"],"award-info":[{"award-number":["R35GM128906"]}],"id":[{"id":"10.13039\/100000057","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000181","name":"United States Air Force Office of Scientific Research","doi-asserted-by":"crossref","award":["FA9550-25-C-B010"],"award-info":[{"award-number":["FA9550-25-C-B010"]}],"id":[{"id":"10.13039\/100000181","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,8,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>A protein\u2019s function is derived from its three-dimensional structure and the motions of the atoms about that structure. The detailed characterization of both macromolecular structure and dynamics provides an opportunity for understanding enzyme catalysis, ligand binding, and allostery, along with providing insights into how the function changes upon mutation or post-translational modification. Among the various methods for characterizing biomolecular motions, nuclear magnetic resonance (NMR) spin relaxation methods are a standard for determining nanosecond global tumbling times along with the amplitude and timescale of faster local motions. Within the model-free formalism, various mathematical models are used to extract dynamic parameters. Unfortunately, as the number of fitted parameters increases within these models, they become mathematically underdetermined for standard NMR relaxation data collected at a single magnetic field, necessitating multi-field datasets. Here, we present SPINDLE, an ensemble of deep neural networks trained on a large synthetic set of NMR relaxation data. Unlike traditional least-squares fitting, SPINDLE predicts both fast and slow timescale dynamics parameters from a single set (i.e., collected at a single magnetic field) of three relaxation datasets using the ensemble for error estimation. We demonstrate a strong correlation to ground truth dynamics parameters on synthetic benchmarks, with more precision than traditional fitting techniques, and precisely reproduce experimental dynamics parameters for \u223c50 proteins with relaxation data in the Biological Magnetic Resonance Data Bank. We also leverage the architecture of the deep neural network to show how the model emphasizes rigid residues for the prediction of global correlation times. This strategy may be useful in the future for elucidating correlated networks of dynamic residues from multiple relaxation datasets.<\/jats:p>","DOI":"10.64898\/2026.07.31.742133","type":"posted-content","created":{"date-parts":[[2026,8,2]],"date-time":"2026-08-02T18:25:23Z","timestamp":1785695123000},"source":"Crossref","is-referenced-by-count":0,"title":["SPINDLE: Unlocking protein dynamics from single-field NMR relaxation data using a deep learning ensemble"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-1077-0006","authenticated-orcid":false,"given":"Olivia E.","family":"Krise","sequence":"first","affiliation":[{"name":"Department of Biochemistry, Molecular Biology and Biophysics, University of Minnesota, Minneapolis, MN, 55455, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2209-5798","authenticated-orcid":false,"given":"Michael P.","family":"Latham","sequence":"additional","affiliation":[{"name":"Department of Biochemistry, Molecular Biology and Biophysics, University of Minnesota, Minneapolis, MN, 55455, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026080418550873000_2026.07.31.742133v1.1","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-021-03819-2"},{"key":"2026080418550873000_2026.07.31.742133v1.2","doi-asserted-by":"publisher","DOI":"10.1126\/science.abj8754"},{"key":"2026080418550873000_2026.07.31.742133v1.3","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1385\/CBB:37:3:187","article-title":"Protein Dynamics from Solution NMR","volume":"37","year":"2002","journal-title":"CBB"},{"key":"2026080418550873000_2026.07.31.742133v1.4","doi-asserted-by":"publisher","DOI":"10.1021\/cr030413t"},{"key":"2026080418550873000_2026.07.31.742133v1.5","doi-asserted-by":"publisher","DOI":"10.1038\/nature04105"},{"key":"2026080418550873000_2026.07.31.742133v1.6","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-018-03562-9"},{"key":"2026080418550873000_2026.07.31.742133v1.7","doi-asserted-by":"publisher","DOI":"10.1126\/science.1130258"},{"key":"2026080418550873000_2026.07.31.742133v1.8","doi-asserted-by":"publisher","DOI":"10.1038\/nature02655"},{"key":"2026080418550873000_2026.07.31.742133v1.9","doi-asserted-by":"publisher","DOI":"10.1021\/ja203686t"},{"key":"2026080418550873000_2026.07.31.742133v1.10","doi-asserted-by":"crossref","first-page":"1639","DOI":"10.1038\/s41598-018-19908-8","article-title":"A dynamic allosteric pathway underlies Rad50 ABC ATPase function in DNA repair","volume":"8","year":"2018","journal-title":"Sci Rep"},{"key":"2026080418550873000_2026.07.31.742133v1.11","doi-asserted-by":"publisher","DOI":"10.1038\/nsmb1132"},{"key":"2026080418550873000_2026.07.31.742133v1.12","doi-asserted-by":"publisher","DOI":"10.1021\/ja00381a009"},{"key":"2026080418550873000_2026.07.31.742133v1.13","doi-asserted-by":"publisher","DOI":"10.1021\/ja00381a010"},{"key":"2026080418550873000_2026.07.31.742133v1.14","doi-asserted-by":"publisher","DOI":"10.1021\/cr040421p"},{"key":"2026080418550873000_2026.07.31.742133v1.15","doi-asserted-by":"publisher","DOI":"10.1021\/ja00168a070"},{"key":"2026080418550873000_2026.07.31.742133v1.16","doi-asserted-by":"publisher","DOI":"10.1006\/jmbi.1994.0073"},{"key":"2026080418550873000_2026.07.31.742133v1.17","doi-asserted-by":"publisher","DOI":"10.1023\/A:1021902006114"},{"key":"2026080418550873000_2026.07.31.742133v1.18","doi-asserted-by":"publisher","DOI":"10.1023\/A:1008305808620"},{"key":"2026080418550873000_2026.07.31.742133v1.19","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btu166"},{"key":"2026080418550873000_2026.07.31.742133v1.20","doi-asserted-by":"publisher","DOI":"10.1023\/A:1023808801134"},{"key":"2026080418550873000_2026.07.31.742133v1.21","doi-asserted-by":"crossref","first-page":"4144","DOI":"10.1038\/s41467-023-39581-4","article-title":"AI-designed NMR spectroscopy RF pulses for fast acquisition at high and ultra-high magnetic fields","volume":"14","year":"2023","journal-title":"Nat Commun"},{"key":"2026080418550873000_2026.07.31.742133v1.22","doi-asserted-by":"crossref","first-page":"pgac133","DOI":"10.1093\/pnasnexus\/pgac133","article-title":"High-fidelity control of spin ensemble dynamics via artificial intelligence: from quantum computing to NMR spectroscopy and imaging","volume":"1","year":"2022","journal-title":"PNAS Nexus"},{"key":"2026080418550873000_2026.07.31.742133v1.23","doi-asserted-by":"crossref","first-page":"1377","DOI":"10.1021\/acs.analchem.0c03087","article-title":"Improvement in Signal-to-Noise Ratio of Liquid-State NMR Spectroscopy via a Deep Neural Network DN-Unet","volume":"93","year":"2021","journal-title":"Anal. Chem"},{"key":"2026080418550873000_2026.07.31.742133v1.24","doi-asserted-by":"crossref","first-page":"5520","DOI":"10.1021\/acs.jpclett.6c00770","article-title":"DEEP Phaser: A Deep Learning Tandem Vision Transformer for Fully Automated NMR Phase Correction","volume":"17","year":"2026","journal-title":"J. Phys. Chem. Lett"},{"key":"2026080418550873000_2026.07.31.742133v1.25","doi-asserted-by":"crossref","first-page":"1245","DOI":"10.1021\/acs.jpclett.5c03529","article-title":"Phase Model-Driven Deep Learning for Robust Phase Correction in High-Throughput NMR-Based Metabolomics","volume":"17","year":"2026","journal-title":"J. Phys. Chem. Lett"},{"key":"2026080418550873000_2026.07.31.742133v1.26","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-25496-5"},{"key":"2026080418550873000_2026.07.31.742133v1.27","doi-asserted-by":"publisher","DOI":"10.1007\/s10858-021-00366-w"},{"key":"2026080418550873000_2026.07.31.742133v1.28","doi-asserted-by":"crossref","first-page":"577","DOI":"10.1007\/s10858-019-00265-1","article-title":"Using Deep Neural Networks to Reconstruct Non-uniformly Sampled NMR Spectra","volume":"73","year":"2019","journal-title":"J Biomol NMR"},{"key":"2026080418550873000_2026.07.31.742133v1.29","doi-asserted-by":"crossref","unstructured":"Zhan, H. et al. Fast Pure Shift NMR Spectroscopy Using Attention-Assisted Deep Neural Network. Advanced Science 11, 2309810 (2024).","DOI":"10.1002\/advs.202309810"},{"key":"2026080418550873000_2026.07.31.742133v1.30","doi-asserted-by":"crossref","first-page":"eadr2155","DOI":"10.1126\/sciadv.adr2155","article-title":"A combined NMR and deep neural network approach for enhancing the spectral resolution of aromatic side chains in proteins","volume":"10","year":"2024","journal-title":"Sci. Adv"},{"key":"2026080418550873000_2026.07.31.742133v1.31","doi-asserted-by":"crossref","unstructured":"Prestegard, J. H. et al. Neural net analysis of NMR spectra from strongly-coupled spin systems. Journal of Magnetic Resonance 368, 107792 (2024).","DOI":"10.1016\/j.jmr.2024.107792"},{"key":"2026080418550873000_2026.07.31.742133v1.32","doi-asserted-by":"crossref","first-page":"e2527937123","DOI":"10.1073\/pnas.2527937123","article-title":"Two-dimensional NMR from a single pulse: Reconstructing heteronuclear 2D spectra via off-resonance decoupling and deep neural networks","volume":"123","year":"2026","journal-title":"Proc. Natl. Acad. Sci. U.S.A"},{"key":"2026080418550873000_2026.07.31.742133v1.33","doi-asserted-by":"publisher","DOI":"10.1007\/s10858-010-9433-9"},{"key":"2026080418550873000_2026.07.31.742133v1.34","doi-asserted-by":"publisher","DOI":"10.1007\/s10858-013-9741-y"},{"key":"2026080418550873000_2026.07.31.742133v1.35","doi-asserted-by":"crossref","unstructured":"Hare, Brian J . & Prestegard, James H . Application of neural networks to automated assignment of NMR spectra of proteins. J Biomol NMR 4, (1994).","DOI":"10.1007\/BF00178334"},{"key":"2026080418550873000_2026.07.31.742133v1.36","doi-asserted-by":"crossref","first-page":"659","DOI":"10.1016\/0022-2364(91)90157-O","article-title":"Identification of 2D 1H NMR antiphase cross peaks using a neural network","volume":"94","year":"1991","journal-title":"Journal of Magnetic Resonance (1969)"},{"key":"2026080418550873000_2026.07.31.742133v1.37","doi-asserted-by":"crossref","first-page":"256","DOI":"10.1016\/0022-2364(92)90260-E","article-title":"An artificial neural network for classifying cross peaks in two-dimensional NMR spectra","volume":"100","year":"1992","journal-title":"Journal of Magnetic Resonance (1969)"},{"key":"2026080418550873000_2026.07.31.742133v1.38","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-33879-5"},{"key":"2026080418550873000_2026.07.31.742133v1.39","doi-asserted-by":"publisher","DOI":"10.1007\/S10858-020-00350-W"},{"key":"2026080418550873000_2026.07.31.742133v1.40","doi-asserted-by":"crossref","unstructured":"d\u2019Auvergne, E. J. & Gooley, P. R. Optimisation of NMR dynamic models I. Minimisation algorithms and their performance within the model-free and Brownian rotational diffusion spaces. J Biomol NMR 40, 107 (2008).","DOI":"10.1007\/s10858-007-9214-2"},{"key":"2026080418550873000_2026.07.31.742133v1.41","doi-asserted-by":"publisher","DOI":"10.1023\/A:1008313319334"},{"key":"2026080418550873000_2026.07.31.742133v1.42","doi-asserted-by":"publisher","DOI":"10.1021\/ja00155a020"},{"key":"2026080418550873000_2026.07.31.742133v1.43","doi-asserted-by":"publisher","DOI":"10.1007\/s10858-006-9007-z"},{"key":"2026080418550873000_2026.07.31.742133v1.44","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1016\/S0022-2836(02)00137-7","article-title":"NMR Studies of the Backbone Flexibility and Structure of Human Growth Hormone: A Comparison of High and Low pH Conformations","volume":"318","year":"2002","journal-title":"Journal of Molecular Biology"},{"key":"2026080418550873000_2026.07.31.742133v1.45","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1023\/A:1008302101116","article-title":"Lipari\u2013Szabo mapping: A graphical approach to Lipari\u2013Szabo analysis of NMR relaxation data using reduced spectral density mapping","volume":"18","year":"2000","journal-title":"J Biomol NMR"},{"key":"2026080418550873000_2026.07.31.742133v1.46","doi-asserted-by":"crossref","unstructured":"Abragam, A. Principles of Nuclear Magnetism. (Oxford University Press (OUP), 1961).","DOI":"10.1119\/1.1937646"},{"key":"2026080418550873000_2026.07.31.742133v1.47","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1603.04467"},{"key":"2026080418550873000_2026.07.31.742133v1.48","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1412.6980"},{"key":"2026080418550873000_2026.07.31.742133v1.49","doi-asserted-by":"publisher","DOI":"10.1016\/j.bpj.2017.03.011"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.07.31.742133","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T01:55:19Z","timestamp":1785894919000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.07.31.742133"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":49,"URL":"https:\/\/doi.org\/10.64898\/2026.07.31.742133","relation":{},"subject":[],"published":{"date-parts":[[2026,8,1]]},"subtype":"preprint"}}