Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Machine-learned potentials for next-generation matter simulations

Nature Materials volume 20, pages 750–761 (2021)Cite this article

Subjects

Abstract

The choice of simulation methods in computational materials science is driven by a fundamental trade-off: bridging large time- and length-scales with highly accurate simulations at an affordable computational cost. Venturing the investigation of complex phenomena on large scales requires fast yet accurate computational methods. We review the emerging field of machine-learned potentials, which promises to reach the accuracy of quantum mechanical computations at a substantially reduced computational cost. This Review will summarize the basic principles of the underlying machine learning methods, the data acquisition process and active learning procedures. We highlight multiple recent applications of machine-learned potentials in various fields, ranging from organic chemistry and biomolecules to inorganic crystal structure predictions and surface science. We furthermore discuss the developments required to promote a broader use of ML potentials, and the possibility of using them to help solve open questions in materials science and facilitate fully computational materials design.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Simulation methods.
Fig. 2: Representation and models.
Fig. 3: Data acquisition.
Fig. 4: Active learning.
Fig. 5: Applications.
Fig. 6: ML potentials as a general-purpose tool.

References

  1. Sanchez-Lengeling, B. & Aspuru-Guzik, A. Inverse molecular design using machine learning: generative models for matter engineering. Science 361, 360–365 (2018).

    Article  CAS  Google Scholar 

  2. Tabor, D. P. et al. Accelerating the discovery of materials for clean energy in the era of smart automation. Nat. Rev. Mater. 3, 5–20 (2018).

    Article  CAS  Google Scholar 

  3. Popova, M., Isayev, O. & Tropsha, A. Deep reinforcement learning for de novo drug design. Sci. Adv. 4, eaap7885 (2018).

    Article  CAS  Google Scholar 

  4. Car, R. & Parrinello, M. Unified approach for molecular dynamics and density-functional theory. Phys. Rev. Lett. 55, 2471–2474 (1985).

    Article  CAS  Google Scholar 

  5. Payne, M. C., Teter, M. P., Allan, D. C., Arias, T. A. & Joannopoulos, J. D. Iterative minimization techniques for ab initio total-energy calculations: molecular dynamics and conjugate gradients. Rev. Mod. Phys. 64, 1045–1097 (1992).

    Article  CAS  Google Scholar 

  6. Karplus, M. & McCammon, J. A. Molecular dynamics simulations of biomolecules. Nat. Struct. Mol. Biol. 9, 646–652 (2002).

    Article  CAS  Google Scholar 

  7. Sugita, Y. & Okamoto, Y. Replica-exchange molecular dynamics method for protein folding. Chem. Phys. Lett. 314, 141–151 (1999).

    Article  CAS  Google Scholar 

  8. Lindorff-Larsen, K., Piana, S., Dror, R. O. & Shaw, D. E. How fast-folding proteins fold. Science 334, 517–520 (2011).

    Article  CAS  Google Scholar 

  9. Piana, S., Lindorff-Larsen, K. & Shaw, D. E. How robust are protein folding simulations with respect to force field parameterization? Biophys. J. 100, L47–49 (2011).

    Article  CAS  Google Scholar 

  10. van Duin, A. C. T., Dasgupta, S., Lorant, F. & Goddard, W. A. ReaxFF: a reactive force field for hydrocarbons. J. Phys. Chem. A 105, 9396–9409 (2001).

    Article  CAS  Google Scholar 

  11. Behler, J. & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Phys. Rev. Lett. 98, 146401 (2007).

    Article  CAS  Google Scholar 

  12. Behler, J. Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations. Phys. Chem. Chem. Phys. 13, 17930–17955 (2011).

    Article  CAS  Google Scholar 

  13. Izvekov, S., Parrinello, M., Burnham, C. J. & Voth, G. A. Effective force fields for condensed phase systems from ab initio molecular dynamics simulation: a new method for force-matching. J. Chem. Phys. 120, 10896–10913 (2004).

    Article  CAS  Google Scholar 

  14. Blank, T. B., Brown, S. D., Calhoun, A. W. & Doren, D. J. Neural network models of potential energy surfaces. J. Chem. Phys. 103, 4129–4137 (1995).

    Article  CAS  Google Scholar 

  15. Ercolessi, F. & Adams, J. B. Interatomic potentials from first-principles calculations. MRS Proc. 291, 31 (1992).

    Article  Google Scholar 

  16. Ischtwan, J. & Collins, M. A. Molecular potential energy surfaces by interpolation. J. Chem. Phys. 100, 8080–8088 (1994).

    Article  CAS  Google Scholar 

  17. Ho, T. & Rabitz, H. A general method for constructing multidimensional molecular potential energy surfaces from ab initio calculations. J. Chem. Phys. 104, 2584–2597 (1996).

    Article  CAS  Google Scholar 

  18. Smith, J. S., Isayev, O. & Roitberg, A. E. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost. Chem. Sci. 8, 3192–3203 (2017).

    Article  CAS  Google Scholar 

  19. Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O. & Walsh, A. Machine learning for molecular and materials science. Nature 559, 547–555 (2018).

    Article  CAS  Google Scholar 

  20. Behler, J. Representing potential energy surfaces by high-dimensional neural network potentials. J. Phys. Condens. Matter 26, 183001 (2014).

    Article  CAS  Google Scholar 

  21. Bernstein, N. et al. Quantifying chemical structure and machine‐learned atomic energies in amorphous and liquid silicon. Angew. Chem. Int. Ed. Engl. 131, 7131–7135 (2019).

    Article  Google Scholar 

  22. Deringer, V. L., Proserpio, D. M., Csányi, G. & Pickard, C. J. Data-driven learning and prediction of inorganic crystal structures. Faraday Discuss. 211, 45–59 (2018).

    Article  CAS  Google Scholar 

  23. Geiger, P. & Dellago, C. Neural networks for local structure detection in polymorphic systems. J. Chem. Phys. 139, 164105 (2013).

    Article  CAS  Google Scholar 

  24. Simm, G. N. & Reiher, M. Error-controlled exploration of chemical reaction networks with Gaussian processes. J. Chem. Theory Comput. 14, 5238–5248 (2018).

    Article  CAS  Google Scholar 

  25. An, S. J. et al. The state of understanding of the lithium-ion-battery graphite solid electrolyte interphase (SEI) and its relationship to formation cycling. Carbon 105, 52–76 (2016).

    Article  CAS  Google Scholar 

  26. Reddy, G., Liu, Z. & Thirumalai, D. Denaturant-dependent folding of GFP. Proc. Natl Acad. Sci. USA 109, 17832–17838 (2012).

    Article  CAS  Google Scholar 

  27. Shu, Y. & Levine, B. G. Communication: non-radiative recombination via conical intersection at a semiconductor defect. J. Chem. Phys. 139, 081102 (2013).

    Article  CAS  Google Scholar 

  28. Ercolessi, F. & Adams, J. B. Interatomic potentials from first-principles calculations: the force-matching method. Europhys. Lett. 26, 583–588 (1994).

    Article  CAS  Google Scholar 

  29. Lorenz, S., Groß, A. & Scheffler, M. Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks. Chem. Phys. Lett. 395, 210–215 (2004).

    Article  CAS  Google Scholar 

  30. Ciresan, D. C., Meier, U., Masci, J., Gambardella, L. M. & Schmidhuber, J. Flexible, high performance convolutional neural networks for image classification. In Twenty-Second International Joint Conference on Artificial Intelligence (IJCAI, 2011).

  31. Wang, L., Wu, W., Xiao, J. & Yi, Y. Large scale artificial neural network training using multi-GPUs. Preprint at https://arxiv.org/abs/1511.04348 (2015).

  32. Steinkraus, D., Buck, I. & Simard, P. Y. Using GPUs for machine learning algorithms. In Eighth International Conference on Document Analysis and Recognition Vol. 2 1115–1120 (IEEE, 2005).

  33. LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015).

    Article  CAS  Google Scholar 

  34. Yao, K., Herr, J. E., Toth, D. W., Mckintyre, R. & Parkhill, J. The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics. Chem. Sci. 9, 2261–2269 (2018).

    Article  CAS  Google Scholar 

  35. Lubbers, N., Smith, J. S. & Barros, K. Hierarchical modeling of molecular energies using a deep neural network. J. Chem. Phys. 148, 241715 (2018).

    Article  CAS  Google Scholar 

  36. Behler, J. Constructing high-dimensional neural network potentials: a tutorial review. Int. J. Quantum Chem. 115, 1032–1050 (2015).

    Article  CAS  Google Scholar 

  37. Smith, J. S., Isayev, O. & Roitberg, A. E. ANI-1, a data set of 20 million calculated off-equilibrium conformations for organic molecules. Sci. Data 4, 170193 (2017).

    Article  CAS  Google Scholar 

  38. Schütt, K. T., Arbabzadah, F., Chmiela, S., Müller, K. R. & Tkatchenko, A. Quantum-chemical insights from deep tensor neural networks. Nat. Commun. 8, 13890 (2017).

    Article  CAS  Google Scholar 

  39. Chmiela, S. et al. Machine learning of accurate energy-conserving molecular force fields. Sci. Adv. 3, e1603015 (2017).

    Article  CAS  Google Scholar 

  40. Behler, J. Perspective: machine learning potentials for atomistic simulations. J. Chem. Phys. 145, 170901 (2016).

    Article  CAS  Google Scholar 

  41. Simm, G. N., Proppe, J. & Reiher, M. Error assessment of computational models in chemistry. Chimia 71, 202–208 (2017).

    Article  CAS  Google Scholar 

  42. Settles, B. Active Learning (Morgan & Claypool, 2012).

  43. Brown, D. F. R., Gibbs, M. N. & Clary, D. C. Combining ab initio computations, neural networks, and diffusion Monte Carlo: an efficient method to treat weakly bound molecules. J. Chem. Phys. 105, 7597–7604 (1996).

    Article  CAS  Google Scholar 

  44. Collins, C. R., Gordon, G. J., von Lilienfeld, O. A. & Yaron, D. J. Constant size descriptors for accurate machine learning models of molecular properties. J. Chem. Phys. 148, 241718 (2018).

    Article  CAS  Google Scholar 

  45. Cubuk, E. D., Malone, B. D., Onat, B., Waterland, A. & Kaxiras, E. Representations in neural network based empirical potentials. J. Chem. Phys. 147, 024104 (2017).

    Article  CAS  Google Scholar 

  46. Faber, F., Lindmaa, A., von Lilienfeld, O. A. & Armiento, R. Crystal structure representations for machine learning models of formation energies. Int. J. Quant. Chem. 115, 1094–1101 (2015).

    Article  CAS  Google Scholar 

  47. Schütt, K. T. et al. How to represent crystal structures for machine learning: towards fast prediction of electronic properties. Phys. Rev. B 89, 205118 (2014).

    Article  CAS  Google Scholar 

  48. Duvenaud, D. K. et al. Convolutional networks on graphs for learning molecular fingerprints. In Advances in Neural Information Processing Systems Vol. 28 2224–2232 (NeurIPS, 2015).

  49. Rupp, M., Tkatchenko, A., Müller, K.-R. & von Lilienfeld, O. A. Fast and accurate modeling of molecular atomization energies with machine learning. Phys. Rev. Lett. 108, 058301 (2012).

    Article  CAS  Google Scholar 

  50. Friederich, P., Konrad, M., Strunk, T. & Wenzel, W. Machine learning of correlated dihedral potentials for atomistic molecular force fields. Sci. Rep. 8, 2559 (2018).

    Article  CAS  Google Scholar 

  51. Westermayr, J. et al. Machine learning enables long time scale molecular photodynamics simulations. Chem. Sci. 10, 8100–8107 (2019).

    Article  CAS  Google Scholar 

  52. Bartók, A. P., Gillan, M. J., Manby, F. R. & Csányi, G. Machine-learning approach for one- and two-body corrections to density functional theory: applications to molecular and condensed water. Phys. Rev. B Condens. Matter 88, 054104 (2013).

    Article  CAS  Google Scholar 

  53. Huo, H. & Rupp, M. Unified representation of molecules and crystals for machine learning. Preprint at https://arxiv.org/abs/1704.06439 (2017).

  54. Wang, H., Zhang, L., Han, J. & Weinan, E. DeePMD-kit: a deep learning package for many-body potential energy representation and molecular dynamics. Comput. Phys. Commun. 228, 178–184 (2018).

    Article  CAS  Google Scholar 

  55. Yoo, D. et al. Atomic energy mapping of neural network potential. Phys. Rev. Mater. 3, 093802 (2019).

    Article  CAS  Google Scholar 

  56. Chapman, D. E., Steck, J. K. & Nerenberg, P. S. Optimizing protein-protein van der Waals interactions for the AMBER ff9x/ff12 force field. J. Chem. Theory Comput. 10, 273–281 (2014).

    Article  CAS  Google Scholar 

  57. Handley, C. M. & Popelier, P. L. A. Dynamically polarizable water potential based on multipole moments trained by machine learning. J. Chem. Theory Comput. 5, 1474–1489 (2009).

    Article  CAS  Google Scholar 

  58. Mills, M. J. L. & Popelier, P. L. A. Electrostatic forces: formulas for the first derivatives of a polarizable, anisotropic electrostatic potential energy function based on machine learning. J. Chem. Theory Comput. 10, 3840–3856 (2014).

    Article  CAS  Google Scholar 

  59. Unke, O. T. & Meuwly, M. PhysNet: a neural network for predicting energies, forces, dipole moments and partial charges. J. Chem. Theory Comput. 15, 3678–3693 (2019).

    Article  CAS  Google Scholar 

  60. Grimme, S. Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J. Comput. Chem. 27, 1787–1799 (2006).

    Article  CAS  Google Scholar 

  61. Gastegger, M., Behler, J. & Marquetand, P. Machine learning molecular dynamics for the simulation of infrared spectra. Chem. Sci. 8, 6924–6935 (2017).

    Article  CAS  Google Scholar 

  62. Bereau, T., Andrienko, D. & von Lilienfeld, O. A. Transferable atomic multipole machine learning models for small organic molecules. J. Chem. Theory Comput. 11, 3225–3233 (2015).

    Article  CAS  Google Scholar 

  63. Sifain, A. E. et al. Discovering a transferable charge assignment model using machine learning. J. Phys. Chem. Lett. 9, 4495–4501 (2018).

    Article  CAS  Google Scholar 

  64. Bereau, T., DiStasio, R. A. Jr, Tkatchenko, A. & von Lilienfeld, O. A. Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning. J. Chem. Phys. 148, 241706 (2018).

    Article  CAS  Google Scholar 

  65. Ramakrishnan, R. & von Lilienfeld, O. A. Many molecular properties from one kernel in chemical space. Chimia 69, 182–186 (2015).

    Article  CAS  Google Scholar 

  66. Rupp, M. Machine learning for quantum mechanics in a nutshell. Int. J. Quantum Chem. 115, 1058–1073 (2015).

    Article  CAS  Google Scholar 

  67. Bartók, A. P., Payne, M. C., Kondor, R. & Csányi, G. Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons. Phys. Rev. Lett. 104, 136403 (2010).

    Article  CAS  Google Scholar 

  68. Deringer, V. L. & Csányi, G. Machine learning based interatomic potential for amorphous carbon. Phys. Rev. B Condens. Matter 95, 094203 (2017).

    Article  Google Scholar 

  69. Thacker, J. C. R. et al. Towards the simulation of biomolecules: optimisation of peptide-capped glycine using FFLUX. Mol. Simula. 44, 881–890 (2018).

    Article  CAS  Google Scholar 

  70. Uteva, E., Graham, R. S., Wilkinson, R. D. & Wheatley, R. J. Interpolation of intermolecular potentials using Gaussian processes. J. Chem. Phys. 147, 161706 (2017).

    Article  CAS  Google Scholar 

  71. Denzel, A. & Kästner, J. Gaussian process regression for transition state search. J. Chem. Theory Comput. 14, 5777–5786 (2018).

    Article  CAS  Google Scholar 

  72. Weiss, Y., Schölkopf, B. & Platt, J. C. (eds) Advances in Neural Information Processing Systems 18: Proceedings of the 2005 Conference (Bradford Books, 2006).

  73. Seeger, M., Williams, C. & Lawrence, N. Fast forward selection to speed up sparse Gaussian process regression. In Artificial Intelligence and Statistics 9 (EPFL, 2003).

  74. Bartók, A. P., Kondor, R. & Csányi, G. On representing chemical environments. Phys. Rev. B 87, 184115 (2013).

    Article  CAS  Google Scholar 

  75. Szlachta, W. J., Bartók, A. P. & Csányi, G. Accuracy and transferability of Gaussian approximation potential models for tungsten. Phys. Rev. B Condens. Matter 90, 104108 (2014).

    Article  CAS  Google Scholar 

  76. Yang, L., Dacek, S. & Ceder, G. Proposed definition of crystal substructure and substructural similarity. Phys. Rev. B 90, 054102 (2014).

    Article  CAS  Google Scholar 

  77. De, S., Bartók, A. P., Csányi, G. & Ceriotti, M. Comparing molecules and solids across structural and alchemical space. Phys. Chem. Chem. Phys. 18, 13754–13769 (2016).

    Article  CAS  Google Scholar 

  78. Bartók, A. P. et al. Machine learning unifies the modeling of materials and molecules. Sci. Adv. 3, e1701816 (2017).

    Article  Google Scholar 

  79. Imbalzano, G. et al. Automatic selection of atomic fingerprints and reference configurations for machine-learning potentials. J. Chem. Phys. 148, 241730 (2018).

    Article  CAS  Google Scholar 

  80. Mahoney, M. W. & Drineas, P. CUR matrix decompositions for improved data analysis. Proc. Natl Acad. Sci. USA 106, 697–702 (2009).

    Article  CAS  Google Scholar 

  81. MacKay, D. J. C. in Models of Neural Networks III (eds Domany, E., van Hemmen, J. L. & Schulten, K.) 211–254 (1996).

  82. Chmiela, S., Sauceda, H. E., Müller, K.-R. & Tkatchenko, A. Towards exact molecular dynamics simulations with machine-learned force fields. Nat. Commun. 9, 3887 (2018).

    Article  CAS  Google Scholar 

  83. Pyzer-Knapp, E. O., Li, K. & Aspuru-Guzik, A. Learning from the Harvard clean energy project: the use of neural networks to accelerate materials discovery. Adv. Funct. Mater. 25, 6495–6502 (2015).

    Article  CAS  Google Scholar 

  84. Faber, F. A., Lindmaa, A., von Lilienfeld, O. A. & Armiento, R. Machine learning energies of 2 million elpasolite (ABC2D6) crystals. Phys. Rev. Lett. 117, 135502 (2016).

    Article  CAS  Google Scholar 

  85. Yao, K., Herr, J. E., Brown, S. N. & Parkhill, J. Intrinsic bond energies from a bonds-in-molecules neural network. J. Phys. Chem. Lett. 8, 2689–2694 (2017).

    Article  CAS  Google Scholar 

  86. Csányi, G., Albaret, T., Payne, M. C. & De Vita, A. ‘Learn on the fly’: a hybrid classical and quantum-mechanical molecular dynamics simulation. Phys. Rev. Lett. 93, 175503 (2004).

    Article  CAS  Google Scholar 

  87. Deringer, V. L. et al. Realistic atomistic structure of amorphous silicon from machine-learning-driven molecular dynamics. J. Phys. Chem. Lett. 9, 2879–2885 (2018).

    Article  CAS  Google Scholar 

  88. Sosso, G. C., Deringer, V. L., Elliott, S. R. & Csányi, G. Understanding the thermal properties of amorphous solids using machine-learning-based interatomic potentials. Mol. Simula. 44, 866–880 (2018).

    Article  CAS  Google Scholar 

  89. Deringer, V. L., Pickard, C. J. & Csányi, G. Data-driven learning of total and local energies in elemental boron. Phys. Rev. Lett. 120, 156001 (2018).

    Article  CAS  Google Scholar 

  90. Botu, V. & Ramprasad, R. Adaptive machine learning framework to accelerate ab initio molecular dynamics. Int. J. Quant. Chem. 115, 1074–1083 (2015).

    Article  CAS  Google Scholar 

  91. Smith, J. S., Nebgen, B., Lubbers, N., Isayev, O. & Roitberg, A. E. Less is more: sampling chemical space with active learning. J. Chem. Phys. 148, 241733 (2018).

    Article  CAS  Google Scholar 

  92. MacKay, D. J. C. Bayesian methods for neural networks: theory and applications. Netw. Comput. Neural Sys. 6, 469–505 (1995).

    Article  Google Scholar 

  93. Proppe, J., Gugler, S. & Reiher, M. Gaussian process-based refinement of dispersion corrections. J. Chem. Theory Comput. 15, 6046–6060 (2019).

    Article  CAS  Google Scholar 

  94. Gómez-Bombarelli, R. et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 4, 268–276 (2018).

    Article  CAS  Google Scholar 

  95. Häse, F., Roch, L. M., Kreisbeck, C. & Aspuru-Guzik, A. Phoenics: a Bayesian optimizer for chemistry. ACS Cent. Sci. 4, 1134–1145 (2018).

    Article  CAS  Google Scholar 

  96. Häse, F., Roch, L. M. & Aspuru-Guzik, A. Chimera: enabling hierarchy based multi-objective optimization for self-driving laboratories. Chem. Sci. 9, 7642–7655 (2018).

    Article  Google Scholar 

  97. Hernández-Lobato, J. M., Requeima, J., Pyzer-Knapp, E. O. & Aspuru-Guzik, A. Parallel and distributed thompson sampling for large-scale accelerated exploration of chemical space. In Proc. 34th Int. Conf. Machine Learning Vol. 70 1470–1479 (JMLR, 2017).

  98. Zhang, L., Wang, H. & E, W. Reinforced dynamics for enhanced sampling in large atomic and molecular systems. J. Chem. Phys. 148, 124113 (2018).

    Article  CAS  Google Scholar 

  99. Gubaev, K., Podryabinkin, E. V. & Shapeev, A. V. Machine learning of molecular properties: locality and active learning. J. Chem. Phys. 148, 241727 (2018).

    Article  CAS  Google Scholar 

  100. Podryabinkin, E. V. & Shapeev, A. V. Active learning of linearly parametrized interatomic potentials. Comput. Mater. Sci. 140, 171–180 (2017).

    Article  CAS  Google Scholar 

  101. Kolb, B., Luo, X., Zhou, X., Jiang, B. & Guo, H. High-dimensional atomistic neural network potentials for molecule-surface interactions: HCl scattering from Au(111). J. Phys. Chem. Lett. 8, 666–672 (2017).

    Article  CAS  Google Scholar 

  102. Quaranta, V., Hellström, M. & Behler, J. Proton-transfer mechanisms at the water-ZnO interface: the role of presolvation. J. Phys. Chem. Lett. 8, 1476–1483 (2017).

    Article  CAS  Google Scholar 

  103. Boes, J. R. & Kitchin, J. R. Neural network predictions of oxygen interactions on a dynamic Pd surface. Mol. Simula. 43, 346–354 (2017).

    Article  CAS  Google Scholar 

  104. Bartók, A. P., Kermode, J., Bernstein, N. & Csányi, G. Machine learning a general-purpose interatomic potential for silicon. Phys. Rev. X 8, 041048 (2018).

    Google Scholar 

  105. Artrith, N., Morawietz, T. & Behler, J. High-dimensional neural-network potentials for multicomponent systems: Applications to zinc oxide. Phys. Rev. B 83, 079114 (2011).

    Article  CAS  Google Scholar 

  106. Mills, K. et al. Extensive deep neural networks for transferring small scale learning to large scale systems. Chem. Sci. 10, 4129–4140 (2019).

    Article  CAS  Google Scholar 

  107. Chiriki, S., Jindal, S. & Bulusu, S. S. Neural network potentials for dynamics and thermodynamics of gold nanoparticles. J. Chem. Phys. 146, 084314 (2017).

    Article  CAS  Google Scholar 

  108. Eckhoff, M. & Behler, J. From molecular fragments to the bulk: development of a neural network potential for MOF-5. J. Chem. Theory Comput. 15, 3793–3809 (2019).

    Article  CAS  Google Scholar 

  109. Caro, M. A., Deringer, V. L., Koskinen, J., Laurila, T. & Csányi, G. Growth mechanism and origin of high sp3 content in tetrahedral amorphous carbon. Phys. Rev. Lett. 120, 166101 (2018).

    Article  CAS  Google Scholar 

  110. Gabardi, S., Sosso, G. G., Behler, J. & Bernasconi, M. Priming effects in the crystallization of the phase change compound GeTe from atomistic simulations. Faraday Discuss. 213, 287–301 (2019).

    Article  CAS  Google Scholar 

  111. Mocanu, F. C. et al. Modeling the phase-change memory material, GeSbTe, with machine-learned interatomic potential. J. Phys. Chem. B 122, 8998–9006 (2018).

    Article  CAS  Google Scholar 

  112. Artrith, N. & Kolpak, A. M. Understanding the composition and activity of electrocatalytic nanoalloys in aqueous solvents: a combination of DFT and accurate neural network potentials. Nano Lett. 14, 2670–2676 (2014).

    Article  CAS  Google Scholar 

  113. Ramakrishnan, R., Dral, P. O., Rupp, M. & von Lilienfeld, O. A. Quantum chemistry structures and properties of 134 kilo molecules. Sci. Data 1, 140022 (2014).

    Article  CAS  Google Scholar 

  114. Fletcher, T. L. & Popelier, P. L. A. Toward amino acid typing for proteins in FFLUX. J. Comput. Chem. 38, 336–345 (2017).

    Article  CAS  Google Scholar 

  115. Blau, S. M., Bennett, D. I. G., Kreisbeck, C., Scholes, G. D. & Aspuru-Guzik, A. Local protein solvation drives direct down-conversion in phycobiliprotein PC645 via incoherent vibronic transport. Proc. Natl Acad. Sci. USA 115, E3342–E3350 (2018).

    Article  CAS  Google Scholar 

  116. Senftle, T. P. et al. The ReaxFF reactive force-field: development, applications and future directions. npj Comput. Mater. 2, 15011 (2016).

    Article  CAS  Google Scholar 

  117. Bolhuis, P. G., Chandler, D., Dellago, C. & Geissler, P. L. Transition path sampling: throwing ropes over rough mountain passes, in the dark. Annu. Rev. Phys. Chem. 53, 291–318 (2002).

    Article  CAS  Google Scholar 

  118. Reiher, M. (ed.) Topics in Current Chemistry: From Quantum Chemistry to Molecular Simulation (Springer, 2007).

  119. Gaus, M., Cui, Q. & Elstner, M. DFTB3: extension of the self-consistent-charge density-functional tight-binding method (SCC-DFTB). J. Chem. Theory Comput. 7, 931–948 (2012).

    Article  CAS  Google Scholar 

  120. Stewart, J. J. P. MOPAC: A semiempirical molecular orbital program. J. Comput. Aided Mol. Des. 4, 1–103 (1990).

    Article  Google Scholar 

  121. Bannwarth, C., Ehlert, S. & Grimme, S. GFN2-xTB-an accurate and broadly parametrized self-consistent tight-binding quantum chemical method with multipole electrostatics and density-dependent dispersion contributions. J. Chem. Theory Comput. 15, 1652–1671 (2019).

    Article  CAS  Google Scholar 

  122. Lin, H. & Truhlar, D. G. QM/MM: what have we learned, where are we, and where do we go from here? Theor. Chem. Acc. 117, 185–199 (2007).

    Article  CAS  Google Scholar 

  123. Senn, H. M. & Thiel, W. QM/MM studies of enzymes. Curr. Opin. Chem. Biol. 11, 182–187 (2007).

    Article  CAS  Google Scholar 

  124. Cui, Q. Perspective: quantum mechanical methods in biochemistry and biophysics. J. Chem. Phys. 145, 140901 (2016).

    Article  CAS  Google Scholar 

  125. Pozun, Z. D. et al. Optimizing transition states via kernel-based machine learning. J. Chem. Phys. 136, 174101 (2012).

    Article  CAS  Google Scholar 

  126. Li, J., Song, K. & Behler, J. A critical comparison of neural network potentials for molecular reaction dynamics with exact permutation symmetry. Phys. Chem. Chem. Phys. 21, 9672–9682 (2019).

    Article  CAS  Google Scholar 

  127. Lu, D. et al. Mode specific dynamics in the H2 + SH → H + H2S reaction. Phys. Chem. Chem. Phys. 18, 29113–29121 (2016).

    Article  CAS  Google Scholar 

  128. Kolb, B., Zhao, B., Li, J., Jiang, B. & Guo, H. Permutation invariant potential energy surfaces for polyatomic reactions using atomistic neural networks. J. Chem. Phys. 144, 224103 (2016).

    Article  CAS  Google Scholar 

  129. Li, J. & Guo, H. Communication: an accurate full 15 dimensional permutationally invariant potential energy surface for the OH + CH4 → H2O + CH3 reaction. J. Chem. Phys. 143, 221103 (2015).

    Article  CAS  Google Scholar 

  130. Gastegger, M. & Marquetand, P. High-dimensional neural network potentials for organic reactions and an improved training algorithm. J. Chem. Theory Comput. 11, 2187–2198 (2015).

    Article  CAS  Google Scholar 

  131. Liu, Q. et al. Constructing high-dimensional neural network potential energy surfaces for gas–surface scattering and reactions. J. Phys. Chem. C. 122, 1761–1769 (2018).

    Article  CAS  Google Scholar 

  132. Del Cueto, M. et al. New perspectives on CO2–Pt(111) interaction with a high-dimensional neural network potential energy surface. J. Phys. Chem. C. 124, 5174–5181 (2020).

    Article  CAS  Google Scholar 

  133. Shakouri, K., Behler, J., Meyer, J. & Kroes, G.-J. Accurate neural network description of surface phonons in reactive gas-surface dynamics: N2 + Ru(0001). J. Phys. Chem. Lett. 8, 2131–2136 (2017).

    Article  CAS  Google Scholar 

  134. Behler, J. First principles neural network potentials for reactive simulations of large molecular and condensed systems. Angew. Chem. Int. Ed. Engl. 56, 12828–12840 (2017).

    Article  CAS  Google Scholar 

  135. Crespo-Otero, R. & Barbatti, M. Recent advances and perspectives on nonadiabatic mixed quantum-classical dynamics. Chem. Rev. 118, 7026–7068 (2018).

    Article  CAS  Google Scholar 

  136. Chen, W.-K., Liu, X.-Y., Fang, W.-H., Dral, P. O. & Cui, G. Deep Learning for nonadiabatic excited-state dynamics. J. Phys. Chem. Lett. 9, 6702–6708 (2018).

    Article  CAS  Google Scholar 

  137. Hu, D., Xie, Y., Li, X., Li, L. & Lan, Z. Inclusion of machine learning kernel ridge regression potential energy surfaces in on-the-fly nonadiabatic molecular dynamics simulation. J. Phys. Chem. Lett. 9, 2725–2732 (2018).

    Article  CAS  Google Scholar 

  138. Häse, F., Kreisbeck, C. & Aspuru-Guzik, A. Machine learning for quantum dynamics: deep learning of excitation energy transfer properties. Chem. Sci. 8, 8419–8426 (2017).

    Article  Google Scholar 

  139. Richings, G. W. & Habershon, S. Direct quantum dynamics using grid-based wave function propagation and machine-learned potential energy surfaces. J. Chem. Theory Comput. 13, 4012–4024 (2017).

    Article  CAS  Google Scholar 

  140. Musil, F., Willatt, M., Langovoy, M. A. & Ceriotti, M. Fast and accurate uncertainty estimation in chemical machine learning. J. Chem. Theory Comput. 15, 906–915 (2019).

    Article  CAS  Google Scholar 

  141. Huan, T. D. et al. Iterative-learning strategy for the development of application-specific atomistic force fields. J. Phys. Chem. C. 123, 20715–20722 (2019).

    Article  CAS  Google Scholar 

  142. Singraber, A., Behler, J. & Dellago, C. Library-based LAMMPS implementation of high-dimensional neural network potentials. J. Chem. Theory Comput. 15, 1827–1840 (2019).

    Article  CAS  Google Scholar 

  143. Schütt, K. T. et al. SchNetPack: a deep learning toolbox for atomistic systems. J. Chem. Theory Comput. 15, 448–455 (2019).

    Article  CAS  Google Scholar 

  144. Häse, F., Valleau, S., Pyzer-Knapp, E. & Aspuru-Guzik, A. Machine learning exciton dynamics. Chem. Sci. 7, 5139–5147 (2016).

    Article  CAS  Google Scholar 

  145. Zhang, L., Wang, H. & E, W. Adaptive coupling of a deep neural network potential to a classical force field. J. Chem. Phys. 149, 154107 (2018).

    Article  CAS  Google Scholar 

  146. Zhang, L., Han, J., Wang, H., Car, R. & E, W. DeePCG: constructing coarse-grained models via deep neural networks. J. Chem. Phys. 149, 034101 (2018).

    Article  CAS  Google Scholar 

  147. Galvelis, R. & Sugita, Y. Neural network and nearest neighbor algorithms for enhancing sampling of molecular dynamics. J. Chem. Theory Comput. 13, 2489–2500 (2017).

    Article  CAS  Google Scholar 

  148. Chiavazzo, E. et al. Intrinsic map dynamics exploration for uncharted effective free-energy landscapes. Proc. Natl Acad. Sci. USA 114, E5494–E5503 (2017).

    Article  CAS  Google Scholar 

  149. Meyer, B., Sawatlon, B., Heinen, S., von Lilienfeld, O. A. & Corminboeuf, C. Machine learning meets volcano plots: computational discovery of cross-coupling catalysts. Chem. Sci. 9, 7069–7077 (2018).

    Article  CAS  Google Scholar 

  150. S Smith, J. et al. Outsmarting quantum chemistry through transfer learning. Preprint at https://doi.org/10.26434/chemrxiv.6744440 (2018).

  151. Schütt, K. T., Sauceda, H. E.-J., Kindermans, P., Tkatchenko, A.-R. & Müller, K. SchNet – a deep learning architecture for molecules and materials. J. Chem. Phys. 148, 241722 (2018).

    Article  CAS  Google Scholar 

  152. Rasmussen, C. E. & Williams, C. K. I. Gaussian Processes for Machine Learning 10th edn (MIT Press, 2006).

  153. Lu, J., Wang, C. & Zhang, Y. Predicting molecular energy using force-field optimized geometries and atomic vector representations learned from an improved deep tensor neural network. J. Chem. Theory Comput. 15, 4113–4121 (2019).

    Article  CAS  Google Scholar 

  154. Nebgen, B. et al. Transferable dynamic molecular charge assignment using deep neural networks. J. Chem. Theory Comput. 14, 4687–4698 (2018).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors thank G. dos Passos Gomes and S. Y. Guo for helpful discussions about the literature in the field of ML potentials. P.F. acknowledges funding by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 795206 (MolDesign). J.P. appreciates funding through an Early Postdoc.Mobility fellowship by the Swiss National Science Foundation (project no. 178463). F.H. acknowledges support from the Herchel Smith Graduate Fellowship and the Jacques-Emile Dubois Student Dissertation Fellowship. The authors acknowledge support by the Canadian Institute for Advanced Research and the Canada 150 Research Chair Program, as well as the generous support of A. G. Frøseth.

Author information

Author notes

  1. Pascal Friederich

    Present address: Institute of Theoretical Informatics, Karlsruhe Institute of Technology, Karlsruhe, Germany

  2. Jonny Proppe

    Present address: Institute of Physical Chemistry, Georg-August University, Göttingen, Germany

Authors and Affiliations

  1. Chemical Physics Theory Group, Department of Chemistry, University of Toronto, Toronto, Ontario, Canada

    Pascal Friederich, Florian Häse, Jonny Proppe & Alán Aspuru-Guzik

  2. Department of Computer Science, University of Toronto, Toronto, Ontario, Canada

    Pascal Friederich, Florian Häse, Jonny Proppe & Alán Aspuru-Guzik

  3. Institute of Nanotechnology, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Germany

    Pascal Friederich

  4. Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada

    Florian Häse & Alán Aspuru-Guzik

  5. Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA

    Florian Häse & Alán Aspuru-Guzik

  6. Lebovic Fellow, Canadian Institute for Advanced Research (CIFAR), Toronto, Ontario, Canada

    Alán Aspuru-Guzik

Authors
  1. Pascal Friederich
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Florian Häse
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jonny Proppe
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Alán Aspuru-Guzik
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Alán Aspuru-Guzik.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Friederich, P., Häse, F., Proppe, J. et al. Machine-learned potentials for next-generation matter simulations. Nat. Mater. 20, 750–761 (2021). https://doi.org/10.1038/s41563-020-0777-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41563-020-0777-6

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing