On the time scales accessible to atomistic numerical modeling, chemical reactions are considered rare events. Therefore, the atomistic simulations are commonly biased along a low-dimensional representation of a chemical reaction in an atomic structure space, i.e., along the collective variables. However, suitable collective variables are often complicated to guess a priori. We propose a novel method of collective variable discovery based on dimensionality reduction of the atomic representation vectors. These linear-scaling and invariant representations can be either fixed (untrained) or learned by supervised training of the end-to-end machine learning potential. The learned representations are expected to reflect not only the structural but also the energetic features of the system that are transferable to all of the reactive transformation covered by the machine learning potential. We demonstrate our approach on four high-barrier reactions ranging from a simple gas-phase hydrogen jump reaction to complex reactions in periodic models of industrially relevant heterogeneous catalysts. High data efficiency, automatized feature extraction, favorable scaling, and retention of inherent invariances are all properties that are expected to enable fast and largely automatic construction of suitable collective variables even in highly complex reactive scenarios such as reactive/catalytic transformations at solid-liquid interfaces.

Department of Physical and Macromolecular Chemistry Faculty of Sciences Charles University Hlavova 8 128 43 Prague 2 Czech Republic

Mathematical Institute Faculty of Mathematics and Physics Charles University Sokolovská 83 186 75 Prague Czech Republic

Zobrazit více v PubMed

Swendsen R. H.; Wang J.-S. Replica Monte Carlo Simulation of Spin-Glasses. Phys. Rev. Lett. 1986, 57, 2607–2609. 10.1103/PhysRevLett.57.2607. PubMed DOI

Sørensen M. R.; Voter A. F. Temperature-accelerated dynamics for simulation of infrequent events. J. Chem. Phys. 2000, 112, 9599–9606. 10.1063/1.481576. DOI

Laio A.; Parrinello M. Escaping free-energy minima. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 12562–12566. 10.1073/pnas.202427399. PubMed DOI PMC

Torrie G.; Valleau J. Nonphysical sampling distributions in Monte Carlo free-energy estimation: Umbrella sampling. J. Comput. Phys. 1977, 23, 187–199. 10.1016/0021-9991(77)90121-8. DOI

Sidky H.; Chen W.; Ferguson A. L. Machine learning for collective variable discovery and enhanced sampling in biomolecular simulation. Mol. Phys. 2020, 118, e1737742.10.1080/00268976.2020.1737742. DOI

Glielmo A.; Husic B. E.; Rodriguez A.; Clementi C.; Noé F.; Laio A. Unsupervised Learning Methods for Molecular Simulation Data. Chem. Rev. 2021, 121, 9722–9758. 10.1021/acs.chemrev.0c01195. PubMed DOI PMC

Sultan M. M.; Pande V. S. Automated design of collective variables using supervised machine learning. J. Chem. Phys. 2018, 149, 094106.10.1063/1.5029972. PubMed DOI

Chen W.; Tan A. R.; Ferguson A. L. Collective variable discovery and enhanced sampling using autoencoders: Innovations in network architecture and error function design. J. Chem. Phys. 2018, 149, 072312.10.1063/1.5023804. PubMed DOI

Belkacemi Z.; Gkeka P.; Leliévre T.; Stoltz G. Chasing Collective Variables Using Autoencoders and Biased Trajectories. J. Chem. Theory Comput. 2022, 18, 59–78. 10.1021/acs.jctc.1c00415. PubMed DOI

Wehmeyer C.; Noé F. Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics. J. Chem. Phys. 2018, 148, 241703.10.1063/1.5011399. PubMed DOI

Duvenaud D.; Maclaurin D.; Aguilera-Iparraguirre J.; Gómez-Bombarelli R.; Hirzel T.; Aspuru-Guzik A.; Adams R. P.. Convolutional Networks on Graphs for Learning Molecular Fingerprints. In 28th Conference on Neural Information Processing Systems (NIPS 2015); Curran Associates, 2015; pp 2224–2232.

Coley C.; Jin W.; Rogers L.; Jamison T. F.; Jaakkola T. S.; Green W. H.; Barzilay R.; Jensen K. F. A graph-convolutional neural network model for the prediction of chemical reactivity. Chem. Sci. 2019, 10, 370–377. 10.1039/C8SC04228D. PubMed DOI PMC

Kensert A.; Bouwmeester R.; Efthymiadis K.; Van Broeck P.; Desmet G.; Cabooter D. Graph Convolutional Networks for Improved Prediction and Interpretability of Chromatographic Retention Data. Anal. Chem. 2021, 93, 15633–15641. 10.1021/acs.analchem.1c02988. PubMed DOI

Schütt K.; Kindermans P. J.; Sauceda H. E. S.; Chmiela S.; Tkatchenko A.; Müller K. R.. SchNet: A continuous-filter convolutional neural network for modeling quantum interactions. In 31st Conference on Neural Information Processing Systems (NIPS 2017); Curran Associates, 2017.

Schütt K. T.; Unke O. T.; Gastegger M. Equivariant message passing for the prediction of tensorial properties and molecular spectra. Proc. Mach. Learn. Res. 2021, 139, 9377–9388.

Batzner S.; Musaelian A.; Sun L.; Geiger M.; Mailoa J. P.; Kornbluth M.; Molinari N.; Smidt T. E.; Kozinsky B. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. 2022, 13, 2453.10.1038/s41467-022-29939-5. PubMed DOI PMC

Kingma D. P.; Welling M.. Auto-Encoding Variational Bayes arXiv (Statistics.Machine Learning), May 1, 2014, 1312.6114, ver. 10. https://arxiv.org/abs/1312.6114.

Cristovao P.; Nakada H.; Tanimura Y.; Asoh H. Generating In-Between Images Through Learned Latent Space Representation Using Variational Autoencoders. IEEE Access 2020, 8, 149456–149467. 10.1109/ACCESS.2020.3016313. DOI

Huang H.; Li Z.; He R.; Sun Z.; Tan T.. IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis. In 32nd Conference on Neural Information Processing Systems (NeurIPS 2018); Curran Associates, 2018.

Gebauer N. W. A.; Gastegger M.; Schütt K.. Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules. In 33rd Conference on Neural Information Processing Systems (NeurIPS 2019); Curran Associates, 2019; pp 7566–7578.

Behler J.; Parrinello M. Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces. Phys. Rev. Lett. 2007, 98, 146401.10.1103/PhysRevLett.98.146401. PubMed DOI

Behler J. Atom-centered symmetry functions for constructing high-dimensional neural network potentials. J. Chem. Phys. 2011, 134, 074106.10.1063/1.3553717. PubMed DOI

Himanen L.; Jäger M. O. J.; Morooka E. V.; Federici Canova F.; Ranawat Y. S.; Gao D. Z.; Rinke P.; Foster A. S. DScribe: Library of descriptors for machine learning in materials science. Comput. Phys. Commun. 2020, 247, 106949.10.1016/j.cpc.2019.106949. DOI

Sun L.; Vandermause J.; Batzner S.; Xie Y.; Clark D.; Chen W.; Kozinsky B. Multitask Machine Learning of Collective Variables for Enhanced Sampling of Rare Events. J. Chem. Theory Comput. 2022, 18, 2341–2353. 10.1021/acs.jctc.1c00143. PubMed DOI

Li Y.; Li L.; Yu J. Applications of Zeolites in Sustainable Chemistry. Chem 2017, 3, 928–949. 10.1016/j.chempr.2017.10.009. DOI

De Wispelaere K.; Ensing B.; Ghysels A.; Meijer E. J.; Van Speybroeck V. Complex Reaction Environments and Competing Reaction Mechanisms in Zeolite Catalysis: Insights from Advanced Molecular Dynamics. Chem. - Eur. J. 2015, 21, 9385–9396. 10.1002/chem.201500473. PubMed DOI

Nielsen M.; Hafreager A.; Brogaard R. Y.; De Wispelaere K.; Falsig H.; Beato P.; Van Speybroeck V.; Svelle S. Collective action of water molecules in zeolite dealumination. Catal. Sci. Technol. 2019, 9, 3721–3725. 10.1039/C9CY00624A. DOI

Heard C. J.; Grajciar L.; Rice C. M.; Pugh S. M.; Nachtigall P.; Ashbrook S. E.; Morris R. E. Fast room temperature lability of aluminosilicate zeolites. Nat. Commun. 2019, 10, 4690.10.1038/s41467-019-12752-y. PubMed DOI PMC

Database of Zeolite Structures. http://www.iza-structure.org/databases/.

Jin M.; Liu M.; Nachtigall P.; Grajciar L.; Heard C. J. Mechanism of Zeolite Hydrolysis under Basic Conditions. Chem. Mater. 2021, 33, 9202–9212. 10.1021/acs.chemmater.1c02799. DOI

Rimola A.; Costa D.; Sodupe M.; Lambert J.-F.; Ugliengo P. Silica Surface Features and Their Role in the Adsorption of Biomolecules: Computational Modeling and Experiments. Chem. Rev. 2013, 113, 4216–4313. 10.1021/cr3003054. PubMed DOI

Jin M.; Veselý O.; Heard C. J.; Kubu M.; Nachtigall P.; Čejka J.; Grajciar L. The Role of Water Loading and Germanium Content in Germanosilicate Hydrolysis. J. Phys. Chem. C 2021, 125, 23744–23757. 10.1021/acs.jpcc.1c06873. DOI

Erlebach A.; Nachtigall P.; Grajciar L. Accurate large-scale simulations of siliceous zeolites by neural network potentials. npj Comput. Mater. 2022, 8, 174.10.1038/s41524-022-00865-w. DOI

Ang S. J.; Wang W.; Schwalbe-Koda D.; Axelrod S.; Gómez-Bombarelli R. Active learning accelerates ab initio molecular dynamics on reactive energy surfaces. Chem 2021, 7, 738–751. 10.1016/j.chempr.2020.12.009. DOI

Saha I.; Erlebach A.; Nachtigall P.; Heard C. J.; Grajciar L. Reactive Neural Network Potential for Aluminosilicate Zeolites and Water: Quantifying the effect of Si/Al ratio on proton solvation and water diffusion in H-FAU. ChemRxiv 2022, 10.26434/chemrxiv-2022-d1sj9. DOI

Promoting transparency and reproducibility in enhanced molecular simulations. Nat. Methods 2019, 16, 670–673. 10.1038/s41592-019-0506-8. PubMed DOI

Tribello G. A.; Bonomi M.; Branduardi D.; Camilloni C.; Bussi G. PLUMED 2: New feathers for an old bird. Phys. Commun. 2014, 185, 604–613. 10.1016/j.cpc.2013.09.018. DOI

Woo T. K.; Margl P. M.; Blöchl P. E.; Ziegler T. A Combined Car-Parrinello QM/MM Implementation for ab Initio Molecular Dynamics Simulations of Extended Systems: Application to Transition Metal Catalysis. J. Phys. Chem. B 1997, 101, 7877–7880. 10.1021/jp9717296. DOI

Grubmüller H.; Heymann B.; Tavan P. Ligand Binding: Molecular Mechanics Calculation of the Streptavidin-Biotin Rupture Force. Science 1996, 271, 997–999. 10.1126/science.271.5251.997. PubMed DOI

Kühne T. D.; Iannuzzi M.; Del Ben M.; Rybkin V. V.; Seewald P.; Stein F.; Laino T.; Khaliullin R. Z.; Schütt O.; Schiffmann F.; Golze D.; Wilhelm J.; Chulkov S.; Bani-Hashemian M. H.; Weber V.; Borštnik U.; Taillefumier M.; Jakobovits A. S.; Lazzaro A.; Pabst H.; Müller T.; Schade R.; Guidon M.; Andermatt S.; Holmberg N.; Schenter G. K.; Hehn A.; Bussy A.; Belleflamme F.; Tabacchi G.; Glöß A.; Lass M.; Bethune I.; Mundy C. J.; Plessl C.; Watkins M.; VandeVondele J.; Krack M.; Hutter J. CP2K: An electronic structure and molecular dynamics software package - Quickstep: Efficient and accurate electronic structure calculations. J. Chem. Phys. 2020, 152, 194103.10.1063/5.0007045. PubMed DOI

Stewart J. Optimization of parameters for semiempirical methods V: Modification of NDDO approximations and application to 70 elements. J. Mol. Model. 2007, 13, 1173–213. 10.1007/s00894-007-0233-4. PubMed DOI PMC

Schütt K. T.; Kessel P.; Gastegger M.; Nicoli K. A.; Tkatchenko A.; Müller K.-R. SchNetPack: A Deep Learning Toolbox For Atomistic Systems. J. Chem. Theory Comput. 2019, 15, 448–455. 10.1021/acs.jctc.8b00908. PubMed DOI

Larsen A. H.; Mortensen J. J.; Blomqvist J.; Castelli I. E.; Christensen R.; Du lak M.; Friis J.; Groves M. N.; Hammer B.; Hargus C.; Hermes E. D.; Jennings P. C.; Jensen P. B.; Kermode J.; Kitchin J. R.; Kolsbjerg E. L.; Kubal J.; Kaasbjerg K.; Lysgaard S.; Maronsson J. B.; Maxson T.; Olsen T.; Pastewka L.; Peterson A.; Rostgaard C.; Schiøtz J.; Scḧutt O.; Strange M.; Thygesen K. S.; Vegge T.; Vilhelmsen L.; Walter M.; Zeng Z.; Jacobsen K. W. The atomic simulation environment—a Python library for working with atoms. J. Phys.: Condens. Matter 2017, 29, 273002.10.1088/1361-648X/aa680e. PubMed DOI

Dietschreit J. C. B.; Diestler D. J.; Ochsenfeld C. How to obtain reaction free energies from free-energy profiles. J. Chem. Phys. 2022, 156, 114105.10.1063/5.0083423. PubMed DOI

Sierka M.; Sauer J. Finding transition structures in extended systems: A strategy based on a combined quantum mechanics-empirical valence bond approach. J. Chem. Phys. 2000, 112, 6983–6996. 10.1063/1.481296. DOI

Chen W.; Ferguson A. L. Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration. J. Chem. Theory Comput. 2018, 39, 2079–2102. 10.1002/jcc.25520. PubMed DOI

Mauro J. C.; Loucks R. J.; Balakrishnan J. A simplified eigenvector-following technique for locating transition points in an energy landscape. J. Phys. Chem. A 2005, 109, 9578–9583. 10.1021/jp053581t. PubMed DOI

Ribeiro J. M. L.; Bravo P.; Wang Y.; Tiwary P. Reweighted autoencoded variational Bayes for enhanced sampling (RAVE). J. Chem. Phys. 2018, 149, 072301.10.1063/1.5025487. PubMed DOI

Rohrdanz M. A.; Zheng W.; Clementi C. Discovering Mountain Passes via Torchlight: Methods for the Definition of Reaction Coordinates and Pathways in Complex Macromolecular Reactions. Annu. Rev. Phys. Chem. 2013, 64, 295–316. 10.1146/annurev-physchem-040412-110006. PubMed DOI

Kresse G.; Hafner J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 1993, 47, 558–561. 10.1103/PhysRevB.47.558. PubMed DOI

Perdew J. P.; Burke K.; Ernzerhof M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865–3868. 10.1103/PhysRevLett.77.3865. PubMed DOI

Blöchl P. E. Projector augmented-wave method. Phys. Rev. B 1994, 50, 17953–17979. 10.1103/PhysRevB.50.17953. PubMed DOI

Kresse G.; Joubert D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 1999, 59, 1758–1775. 10.1103/PhysRevB.59.1758. DOI

Grimme S.; Antony J.; Ehrlich S.; Krieg H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, 132, 154104.10.1063/1.3382344. PubMed DOI

Grimme S.; Ehrlich S.; Goerigk L. Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 2011, 32, 1456–1465. 10.1002/jcc.21759. PubMed DOI

A reactive neural network framework for water-loaded acidic zeolites