{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,11]],"date-time":"2025-09-11T20:13:17Z","timestamp":1757621597436,"version":"3.44.0"},"reference-count":37,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T00:00:00Z","timestamp":1754265600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T00:00:00Z","timestamp":1754265600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"publisher","award":["197629","197629"],"award-info":[{"award-number":["197629","197629"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100008375","name":"University of Basel","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100008375","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Cheminform"],"abstract":"<jats:sec>\n            <jats:title>Abstract<\/jats:title>\n            <jats:p>Generating accurate molecular conformations hinges on sampling effectively from a high-dimensional space of atomic arrangements, which grows exponentially with system size. To ensure physically valid geometries and increase the likelihood of reaching low-energy conformations, it is us ful\u00a0to incorporate prior physicsbased information by recasting them as geometric constraints that naturally arise as nonlinear constraint satisfaction problems. In this work, we propose an approach to embed these strict constraints into neural differential equations, leveraging the denoising diffusion framework. By projecting the stochastic generative dynamics onto a manifold defined by constraint sets, our method enforces exact feasibility at each step, unlike alternative approaches that merely impose soft constraints through probabilistic guidance. This technique generates lower-energy molecular conformations, enables more efficient subspace exploration, and formally subsumes classifier-guidance-type methods by treating geometric constraints as strict algebraic conditions within the diffusion process.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Scientific Contribution<\/jats:title>\n            <jats:p>Neural SHAKE formulates exact manifold\u2011projected score\u2011based diffusion : each reverse-SDEincrement is orthogonally projected, via a Lagrange-multiplier solve, onto the constraint surface\u03c3\u2090(x)=0 for a = 1,\u2026, A, with A the number of independent constraints and thus the manifold\u2019scodimension . This projection preserves global SE(3) symmetry and enforces constraints tosolver tolerance. It induces a well-posed surface Fokker\u2013Planck flow on the (3 N \u2212 A)-dimensional manifold, while a coarea\/Fixman Jacobian carries the ambient 3 N-dimensionaldensity to a normalized density on that manifold, preserving probability mass after the dimensionality reduction.<\/jats:p>\n          <\/jats:sec>","DOI":"10.1186\/s13321-025-01053-w","type":"journal-article","created":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T17:30:03Z","timestamp":1754328603000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Neural SHAKE: geometric constraints in neural differential equations"],"prefix":"10.1186","volume":"17","author":[{"given":"Justin S.","family":"Diamond","sequence":"first","affiliation":[]},{"given":"Markus A.","family":"Lill","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,8,4]]},"reference":[{"key":"1053_CR1","doi-asserted-by":"publisher","first-page":"6457","DOI":"10.1126\/science.aaw1147","volume":"365","author":"Frank No\u00e9","year":"2019","unstructured":"No\u00e9 Frank et al (2019) Boltzmann generators: sampling equilibrium states of many-body systems with deep learning. Science 365:6457. https:\/\/doi.org\/10.1126\/science.aaw1147. (issn: 0036-8075)","journal-title":"Science"},{"key":"1053_CR2","unstructured":"Emiel Hoogeboom et al. (2023) Equivariant Diffusion for Molecule Generation in 3D. In: International Conference on Machine Learning. , pp. 8867\u20138887. arXiv: org\/pdf\/2203.17003pdf"},{"key":"1053_CR3","unstructured":"Hoffmann M, No\u00e9. F(2019) \u201cGenerating Valid Euclidean Distance Matrices\u201d. In: arXiv preprint  (2019). arXiv: org\/abs\/1910.03131"},{"key":"1053_CR4","unstructured":"Gabriele Corso et al (2023). DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking. In: International Conference on Learning Representations. . arXiv: org\/abs\/2210.01776"},{"key":"1053_CR5","unstructured":"Herbert Goldstein, Charles Poole, and John Safko (2002). Classical Mechanics. 3rd. San Francisco, CA: Addison-Wesley, . isbn: 978-0201657029"},{"key":"1053_CR6","unstructured":"Ricky T. Q. Chen et al (2018). Neural Ordinary Differential Equations . In: arXiv preprint  . arXiv: org\/abs\/1806.07366"},{"key":"1053_CR7","unstructured":"Jonathan Ho, Ajay Jain, and Pieter Abbeel. (2020) Denoising Diffusion Probabilistic Models. In: arXiv preprint  . arxiv: org\/abs\/2006.11239"},{"issue":"12","key":"1053_CR8","doi-asserted-by":"publisher","first-page":"5637","DOI":"10.1021\/acs.jcim.0c01015","volume":"60","author":"Maxime Langevin","year":"2020","unstructured":"Langevin Maxime et al (2020) Scaffold-constrained molecular generation. J Chem Inf Model 60(12):5637\u20135646. https:\/\/doi.org\/10.1021\/acs.jcim.0c01015. (Publication Date: December 10, 2020)","journal-title":"J Chem Inf Model"},{"key":"1053_CR9","doi-asserted-by":"publisher","first-page":"914","DOI":"10.1038\/s42256-021-00403-1","volume":"3","author":"Jun Wang","year":"2021","unstructured":"Wang Jun, Hsieh Chia-Yu, Wang Meng et al (2021) Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning. Nat Machine Intell 3:914\u2013922. https:\/\/doi.org\/10.1038\/s42256-021-00403-1","journal-title":"Nat Machine Intell"},{"key":"1053_CR10","unstructured":"Nic Fishman et al. Diffusion models for constrained domains . In: arXiv preprint  arXiv: 2304.05364 (2023)"},{"key":"1053_CR11","volume-title":"Reflected diffusion models","author":"Aaron Lou and Stefano Ermon","year":"2023","unstructured":"Aaron Lou and Stefano Ermon (2023) Reflected diffusion models. PMLR, London"},{"key":"1053_CR12","first-page":"24240","volume":"36","author":"Nic Fishman","year":"2024","unstructured":"Fishman Nic et al (2024) Metropolis sampling for constrained diffusion models. Adv Neural Inf Proc Syst 36:24240","journal-title":"Adv Neural Inf Proc Syst"},{"issue":"4","key":"1053_CR13","doi-asserted-by":"publisher","first-page":"510","DOI":"10.1002\/jcc.10416","volume":"25","author":"A. Coutsias Evangelos","year":"2004","unstructured":"Evangelos A. Coutsias et al (2004) A kinematic view of loop closure. J Comput Chem 25(4):510\u2013528","journal-title":"J Comput Chem"},{"issue":"1","key":"1053_CR14","doi-asserted-by":"publisher","first-page":"176","DOI":"10.1002\/qua.20751","volume":"106","author":"Evangelos A Coutsias","year":"2006","unstructured":"Coutsias Evangelos A et al (2006) Resultants and loop closure. Int J Quantum Chem 106(1):176\u2013189","journal-title":"Int J Quantum Chem"},{"key":"1053_CR15","first-page":"24240","volume":"35","author":"Bowen Jing","year":"2022","unstructured":"Jing Bowen et al (2022) Torsional diffusion for molecular conformer generation. Adv Neural Inf Proc Syst 35:24240","journal-title":"Adv Neural Inf Proc Syst"},{"key":"1053_CR16","unstructured":"Yangtian Zhang et al (2024) DiffPack: A Torsional Diffusion Model for Autoregressive Protein Side-Chain Packing . In: arXiv preprint arXiv:2306.01794 . to appear in NeurIPS 2024"},{"issue":"11","key":"1053_CR17","doi-asserted-by":"publisher","first-page":"1094","DOI":"10.1002\/jcc.27067","volume":"44","author":"Timoth\u00e9e J O\u2019Donnell","year":"2023","unstructured":"O\u2019Donnell Timoth\u00e9e J, Cazals Fr\u00e9d\u00e9ric (2023) Enhanced conformational exploration of protein loops using a global parameterization of the backbone geometry. J Comput Chem 44(11):1094\u20131104","journal-title":"J Comput Chem"},{"issue":"13","key":"1053_CR18","first-page":"1249","volume":"44","author":"Timoth\u00e9e J O\u2019Donnell","year":"2023","unstructured":"O\u2019Donnell Timoth\u00e9e J, Agashe Viraj, Cazals Fr\u00e9d\u00e9ric (2023) Geometric constraints within tripeptides and the existence of tripeptide reconstructions. J Comput Chem 44(13):1249\u20131256","journal-title":"J Comput Chem"},{"key":"1053_CR19","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1140\/epjst\/e2011-01525-9","volume":"200","author":"R Elber","year":"2011","unstructured":"Elber R, Ruymgaart AP, Hess B (2011) SHAKE Parallelization. Euro Phys J Spe Top 200:211\u2013223","journal-title":"Euro Phys J Spe Top"},{"key":"1053_CR20","unstructured":"David, Wales Energy Landscapes Applications to Clusters Biomolecules and Glasses Cambridge University Press"},{"key":"1053_CR21","doi-asserted-by":"crossref","unstructured":"Christine\u2010Andrea, Roth Tom, Dreyfus Charles H., Robert Fr\u00e9d\u00e9ric, Cazals (2016) Hybridizing rapidly exploring random trees and basin hopping yields an improved exploration of energy landscapes Journal of Computational Chemistry 37(8) 739-752 10.1002\/jcc.v37.8 10.1002\/jcc.24256","DOI":"10.1002\/jcc.24256"},{"key":"1053_CR22","unstructured":"Tony, Leli\u00e8vre Mathias, Rousset Gabriel, Stoltz Free Energy Computations A Mathematical Perspective IMPERIAL COLLEGE PRESS"},{"key":"1053_CR23","volume-title":"Probabilistic machine learning: advanced topics","author":"Kevin P Murphy","year":"2023","unstructured":"Murphy Kevin P (2023) Probabilistic machine learning: advanced topics. MIT Press, London"},{"key":"1053_CR24","volume-title":"Neural SHAKE: geometric constraints in graph generative models","author":"Justin Diamond","year":"2024","unstructured":"Diamond Justin, Lill Markus A (2024) Neural SHAKE: geometric constraints in graph generative models. Springer, Cham"},{"key":"1053_CR25","unstructured":"Jascha Sohl-Dickstein et al (2015). Deep Unsupervised Learning Using Nonequilibrium Thermodynamics. In: Proceedings of the 32nd International Conference on Machine Learning (ICML 2015). Vol. 37. Proceedings of Machine Learning Research. PMLR, pp. 2256\u20132265"},{"key":"1053_CR26","unstructured":"Yang Song et al (2021). Score-based Generative Modeling through Stochastic Differential Equations. In: Proceedings of the 9th International Conference on Learning Representations (ICLR 2021). arXiv:2011.13456"},{"issue":"3","key":"1053_CR27","doi-asserted-by":"publisher","first-page":"327","DOI":"10.1016\/0021-9991(77)90098-5","volume":"23","author":"Jean-Paul Ryckaert","year":"1977","unstructured":"Ryckaert Jean-Paul, Ciccotti Giovanni, Berendsen Herman J. C (1977) Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-Alkanes. J Comput Phys 23(3):327\u2013341","journal-title":"J Comput Phys"},{"key":"1053_CR28","doi-asserted-by":"publisher","DOI":"10.4171\/017","author":"Peter Kunkel and Volker Mehrmann","year":"2006","unstructured":"Peter Kunkel and Volker Mehrmann (2006) Differential-algebraic equations: analysis and numerical solution. Euro Math Soc. https:\/\/doi.org\/10.4171\/017","journal-title":"Euro Math Soc"},{"key":"1053_CR29","volume-title":"Introduction to Smooth Manifolds","author":"ML John","year":"2013","unstructured":"John ML (2013) Introduction to Smooth Manifolds. Springer, Cham"},{"key":"1053_CR30","unstructured":"Tianjun Zhang et al (2019) ANODEV2: a coupled neural ODE framework . Adv Neural Inf Proc Syst. 32"},{"key":"1053_CR31","unstructured":"Minkai Xu et al (2022). \u201cGeodiff: A geometric diffusion model for molecular conformation generation\u201d. In: arXiv preprint  . arxiv: org\/abs\/2203.02923"},{"key":"1053_CR32","doi-asserted-by":"publisher","first-page":"1265","DOI":"10.1093\/nar\/gkad976","volume":"52","author":"C Knox","year":"2024","unstructured":"Knox C, Wilson M, Klinger CM et al (2024) DrugBank 6.0: the drugbank knowledgebase for 2024. Nucleic Acids Res 52:1265\u20131275. https:\/\/doi.org\/10.1093\/nar\/gkad976","journal-title":"Nucleic Acids Res"},{"key":"1053_CR33","unstructured":"Greg Landrum (2016). RDKit: Open-Source Cheminformatics Software. https:\/\/www.rdkit.org. Release 2016_09_4."},{"key":"1053_CR34","volume-title":"Theory and Applications of Distance Geometry","author":"Leonard M Blumenthal","year":"1953","unstructured":"Blumenthal Leonard M (1953) Theory and Applications of Distance Geometry. Clarendon Press, Oxford"},{"issue":"3","key":"1053_CR35","doi-asserted-by":"publisher","first-page":"449","DOI":"10.1016\/0021-9991(78)90081-5","volume":"26","author":"GM Crippen","year":"1978","unstructured":"Crippen GM (1978) Rapid calculation of coordinates from distance matrices. J Comput Phys 26(3):449\u2013452","journal-title":"J Comput Phys"},{"issue":"18","key":"1053_CR36","doi-asserted-by":"publisher","first-page":"7737","DOI":"10.1063\/1.477419","volume":"109","author":"Michiel Sprik","year":"1998","unstructured":"Sprik Michiel, Ciccotti Giovanni (1998) Free energy from constrained molecular dynamics. J Chem Phys 109(18):7737\u20137744","journal-title":"J Chem Phys"},{"issue":"8","key":"1053_CR37","doi-asserted-by":"publisher","first-page":"3050","DOI":"10.1073\/pnas.71.8.3050","volume":"71","author":"Marshall Fixman","year":"1974","unstructured":"Fixman Marshall (1974) Classical statistical mechanics of constraints: a theorem and application to polymers. Proc Natl Acad Sci USA 71(8):3050\u20133053","journal-title":"Proc Natl Acad Sci USA"}],"container-title":["Journal of Cheminformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-025-01053-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13321-025-01053-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-025-01053-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T14:17:57Z","timestamp":1757341077000},"score":1,"resource":{"primary":{"URL":"https:\/\/jcheminf.biomedcentral.com\/articles\/10.1186\/s13321-025-01053-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,4]]},"references-count":37,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1053"],"URL":"https:\/\/doi.org\/10.1186\/s13321-025-01053-w","relation":{},"ISSN":["1758-2946"],"issn-type":[{"type":"electronic","value":"1758-2946"}],"subject":[],"published":{"date-parts":[[2025,8,4]]},"assertion":[{"value":"12 February 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 July 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 August 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"115"}}