Hostname: page-component-586b7cd67f-t7fkt Total loading time: 0 Render date: 2024-11-30T07:34:06.876Z Has data issue: false hasContentIssue false
  • English
  • Français

Machine learning prediction of accurate atomization energies of organic molecules from low-fidelity quantum chemical calculations

Published online by Cambridge University Press:  27 August 2019

Logan Ward Open the ORCID record for Logan Ward [Opens in a new window] ,
Ben Blaiszik Open the ORCID record for Ben Blaiszik [Opens in a new window] ,
Ian Foster Open the ORCID record for Ian Foster [Opens in a new window] ,
Rajeev S. Assary ,
Badri Narayanan  and
Larry Curtiss
Logan Ward*
Affiliation:
Data Science and Learning Division, Argonne National Laboratory, Lemont, IL, USA Department of Computer Science, University of Chicago, Chicago, IL, USA
Ben Blaiszik
Affiliation:
Data Science and Learning Division, Argonne National Laboratory, Lemont, IL, USA Globus, University of Chicago, Chicago, IL, USA
Ian Foster
Affiliation:
Data Science and Learning Division, Argonne National Laboratory, Lemont, IL, USA Department of Computer Science, University of Chicago, Chicago, IL, USA Globus, University of Chicago, Chicago, IL, USA
Rajeev S. Assary
Affiliation:
Joint Center for Energy Storage Research (JCESR), Argonne National Laboratory, Lemont, IL, USA Materials Science Division, Argonne National Laboratory, Lemont, IL, USA
Badri Narayanan
Affiliation:
Materials Science Division, Argonne National Laboratory, Lemont, IL, USA Department of Mechanical Engineering, University of Louisville, Louisville, KY, USA
Larry Curtiss
Affiliation:
Joint Center for Energy Storage Research (JCESR), Argonne National Laboratory, Lemont, IL, USA Materials Science Division, Argonne National Laboratory, Lemont, IL, USA
*
Address all correspondence to Logan Ward at [email protected]
Get access

Abstract

Recent studies illustrate how machine learning (ML) can be used to bypass a core challenge of molecular modeling: the trade-off between accuracy and computational cost. Here, we assess multiple ML approaches for predicting the atomization energy of organic molecules. Our resulting models learn the difference between low-fidelity, B3LYP, and high-accuracy, G4MP2, atomization energies and predict the G4MP2 atomization energy to 0.005 eV (mean absolute error) for molecules with less than nine heavy atoms (training set of 117,232 entries, test set 13,026) and 0.012 eV for a small set of 66 molecules with between 10 and 14 heavy atoms. Our two best models, which have different accuracy/speed trade-offs, enable the efficient prediction of G4MP2-level energies for large molecules and are available through a simple web interface.

Type
Artificial Intelligence Research Letters
Information
MRS Communications , Volume 9 , Issue 3 , September 2019 , pp. 891 - 899
Copyright
Copyright © The Author(s) 2019 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1.Curtiss, L.A., Redfern, P.C., and Raghavachari, K.: Gn theory. Wiley Interdiscip. Rev. Comput. Mol. Sci. 1, 810–825 (2011).Google Scholar
2.Curtiss, L.A., Redfern, P.C., and Raghavachari, K.: Gaussian-4 theory using reduced order perturbation theory. J. Chem. Phys. 127, 124105 (2007).Google Scholar
3.Mardirossian, N. and Head-Gordon, M.: Thirty years of density functional theory in computational chemistry: an overview and extensive assessment of 200 density functionals. Mol. Phys. 115, 2315–2372 (2017).Google Scholar
4.Becke, A.D.: A new mixing of Hartree–Fock and local density-functional theories. J. Chem. Phys. 98, 1372 (1993).Google Scholar
5.Ward, L. and Wolverton, C.: Atomistic calculations and materials informatics: a review. Curr. Opin. Solid State Mater. Sci. 21, 167–176 (2017).Google Scholar
6.Handley, C.M. and Behler, J.: Next generation interatomic potentials for condensed systems. Eur. Phys. J. B 87, 152 (2014).Google Scholar
7.Rupp, M.: Machine learning for quantum mechanics in a nutshell. Int. J. Quantum Chem. 115, 1058–1073 (2015).Google Scholar
8.Ramakrishnan, R., Dral, P.O., Rupp, M., and von Lilienfeld, O.A.: Big data meets quantum chemistry approximations: the Δ-machine learning approach. J. Chem. Theory Comput. 11, 2087–2096 (2015).Google Scholar
9.Zaspel, P., Huang, B., Harbrecht, H., and von Lilienfeld, O.A.: Boosting quantum machine learning models with a multilevel combination technique: pople diagrams revisited. J. Chem. Theory Comput. 15, 1546–1559 (2019).Google Scholar
10.Pilania, G., Gubernatis, J.E., and Lookman, T.: Multi-fidelity machine learning models for accurate bandgap predictions of solids. Comput. Mater. Sci. 129, 156–163 (2017).Google Scholar
11.Seko, A., Maekawa, T., Tsuda, K., and Tanaka, I.: Machine learning with systematic density-functional theory calculations: application to melting temperatures of single- and binary-component solids. Phys. Rev. B 89, 054303 (2014).Google Scholar
12.Smith, J.S., Nebgen, B.T., Zubatyuk, R., Lubbers, N., Devereux, C., Barros, K., Tretiak, S., Isayev, O., and Roitbert, A.E.: Outsmarting quantum chemistry through transfer learning universal neural network potentials for organic molecules. ChemArXiv (2018). 10.26434/chemrxiv.6744440.Google Scholar
13.Schütt, K.T., Sauceda, H.E., Kindermans, P.-J., Tkatchenko, A., and Müller, K.-R.: Schnet – a deep learning architecture for molecules and materials. J. Chem. Phys. 148, 241722 (2018).Google Scholar
14.Faber, F.A., Christensen, A.S., Huang, B., and von Lilienfeld, O.A.: Alchemical and structural distribution based representation for universal quantum machine learning. J. Chem. Phys. 148, 241717 (2018).Google Scholar
15.Chard, R., Li, Z., Chard, K., Ward, L., Babuji, Y., Woodard, A., Tuecke, S., Blaiszik, B., Franklin, M.J., and Foster, I.: DLHub: Model and Data Serving for Science (Cornell University, 2018). https://arxiv.org/abs/1811.11213Google Scholar
16.Narayanan, B., Redfern, P.C., Assary, R.S., and Curtiss, L.A.: Accurate quantum chemical energies for 133 000 organic molecules. Chem. Sci. (2019). doi:10.1039/C9SC02834JGoogle Scholar
17.Ramakrishnan, R., Dral, P.O., Rupp, M., and von Lilienfeld, O.A.: Quantum chemistry structures and properties of 134 kilo molecules. Sci. Data 1, 140022 (2014).Google Scholar
18.Blaiszik, B., Chard, K., Pruyne, J., Ananthakrishnan, R., Tuecke, S., and Foster, I.: The materials data facility: data services to advance materials science research. JOM 68, 2045–2052 (2016).Google Scholar
19.Ward, L., Blaiszik, B., Foster, I., Assary, R.S., Narayanan, B., and Curtiss, L.A.: Dataset for Machine Learning Prediction of Accurate Atomization Energies of Organic Molecules from Low-Fidelity Quantum Chemical Calculations (Materials Data Facility, 2019). doi:10.18126/M2V65ZGoogle Scholar
20.https://github.com/globus-labs/g4mp2-atomization-energy.Google Scholar
21.Gilmer, J., Schoenholz, S.S., Riley, P.F., Vinyals, O., and Dahl, G.E.: Neural Message Passing for Quantum Chemistry (2017). http://arxiv.org/abs/1704.01212.Google Scholar
22.Wu, Z., Ramsundar, B., Feinberg, E.N., Gomes, J., Geniesse, C., Pappu, A.S., Leswing, K., and Pande, V.: Moleculenet: a benchmark for molecular machine learning. Chem. Sci. 9, 513–530 (2018).Google Scholar
23.Paul, A., Jha, D., Al-Bahrani, R., Liao, W., Choudhary, A., and Agrawal, A.: CheMixNet: Mixed DNN Architectures for Predicting Chemical Properties Using Multiple Molecular Representations (2018). http://arxiv.org/abs/1811.08283.Google Scholar
24.Schütt, K.T., Kessel, P., Gastegger, M., Nicoli, K.A., Tkatchenko, A., and Müller, K.-R.: Schnetpack: a deep learning toolbox for atomistic systems. J. Chem. Theory Comput. 15, 448–455 (2019).Google Scholar
25.Huang, B. and von Lilienfeld, O.A.: The “DNA” of Chemistry: Scalable Quantum Machine Learning with “Amons”, 2017 http://arxiv.org/abs/1707.04146.Google Scholar
26.Christensen, A.S., Faber, F.A., Huang, B., Bratholm, L.A., Tkatchenko, A., Müller, K.-R., and von Lilienfeld, O.A.: qmlcode/qml: Release v0.3.1 (2017). doi:10.5281/ZENODO.817332.Google Scholar
27.Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.: Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011).Google Scholar
28.Baxter, J.: A Bayesian Information theoretic model of learning to learn via multiple task sampling. Mach. Learn 28, 7–39 (1997).Google Scholar
29.Browning, N.J., Ramakrishnan, R., von Lilienfeld, O.A., and Roethlisberger, U.: Genetic optimization of training sets for improved machine learning models of molecular properties. J. Phys. Chem. Lett. 8, 1351–1359 (2017).Google Scholar
30.Hy, T.S., Trivedi, S., Pan, H., Anderson, B.M., and Kondor, R.: Predicting molecular properties with covariant compositional networks. J. Chem. Phys. 148 (2018).Google Scholar
31.Kearnes, S., McCloskey, K., Berndl, M., Pande, V., and Riley, P.: Molecular graph convolutions: moving beyond fingerprints. J. Comput. Aided Mol. Des. 30, 595–608 (2016).Google Scholar
32.Coley, C.W., Jin, W., Rogers, L., Jamison, T.F., Jaakkola, T.S., Green, W.H., Barzilay, R., and Jensen, K.F.: A graph-convolutional neural network model for the prediction of chemical reactivity. Chem. Sci. 10, 370–377 (2019).Google Scholar
33.Halgren, T.A.: Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94. J. Comput. Chem. 17, 490–519 (1996).Google Scholar
34.O'Boyle, N.M., Banck, M., James, C.A., Morley, C., Vandermeersch, T., and Hutchison, G.R.: Open babel: an open chemical toolbox. J. Cheminform. 3, 33 (2011).Google Scholar
35.Gebauer, N.W.A., Gastegger, M., and Schütt, K.T.: Generating Equilibrium Molecules with Deep Neural Networks (2018). http://arxiv.org/abs/1810.11347.Google Scholar
36.Yao, K., Herr, J.E., Toth, D.W., Mckintyre, R., and Parkhill, J.: The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics. Chem. Sci. 9, 2261–2269 (2018).Google Scholar
37.Nakata, M., Shimazaki, T., Hashimoto, M., and Maeda, T.: PubChemQC PM6: A Dataset of 221 Million Molecules with Optimized Molecular Geometries and Electronic Properties (2019) pp. 1–33. http://arxiv.org/abs/1904.06046.Google Scholar
38.Towns, J., Cockerill, T., Dahan, M., Foster, I., Gaither, K., Grimshaw, A., Hazlewood, V., Lathrop, S., Lifka, D., Peterson, G.D., Roskies, R., Scott, J.R., and Wilkens-Diehr, N.: XSEDE: accelerating scientific discovery. Comput. Sci. Eng. 16, 62–74 (2014).Google Scholar
39.Stewart, C.A., Turner, G., Vaughn, M., Gaffney, N.I., Cockerill, T.M., Foster, I., Hancock, D., Merchant, N., Skidmore, E., Stanzione, D., Taylor, J., and Tuecke, S.: Jetstream: a self-provisioned, scalable science and engineering cloud environment. In Proc. 2015 XSEDE Conf. Sci. Adv. Enabled by Enhanc. Cyberinfrastructure - XSEDE ’15; ACM Press, New York, NY, USA, 2015; pp. 1–8.Google Scholar