Display options
Cite
Davie SJ, Di Pasquale N, Popelier PL. Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer. J Comput Chem. 2016;37(27):2409-22doi: 10.1002/jcc.24465.
Davie, S. J., Di Pasquale, N., & Popelier, P. L. (2016). Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer. Journal of computational chemistry, 37(27), 2409-22. https://doi.org/10.1002/jcc.24465
Davie, Stuart J, et al. "Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer." Journal of computational chemistry vol. 37,27 (2016): 2409-22. doi: https://doi.org/10.1002/jcc.24465
Davie SJ, Di Pasquale N, Popelier PL. Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer. J Comput Chem. 2016 Oct 15;37(27):2409-22. doi: 10.1002/jcc.24465. Epub 2016 Aug 18. PMID: 27535711; PMCID: PMC5031213.
Copy Download .nbib Format: NLM
Share it on

J Comput Chem. 2016 Oct 15;37(27):2409-22. doi: 10.1002/jcc.24465. Epub 2016 Aug 18.

Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer.

Journal of computational chemistry

Stuart J Davie, Nicodemo Di Pasquale, Paul L A Popelier

Affiliations

  1. Manchester Institute of Biotechnology (MIB), 131 Princess Street, Manchester M1 7DN, Great Britain and School of Chemistry, University of Manchester, Oxford Road, Manchester, M13 9PL, Great Britain.
  2. Manchester Institute of Biotechnology (MIB), 131 Princess Street, Manchester M1 7DN, Great Britain and School of Chemistry, University of Manchester, Oxford Road, Manchester, M13 9PL, Great Britain. [email protected].

PMID: 27535711 PMCID: PMC5031213 DOI: 10.1002/jcc.24465

Abstract

Machine learning algorithms have been demonstrated to predict atomistic properties approaching the accuracy of quantum chemical calculations at significantly less computational cost. Difficulties arise, however, when attempting to apply these techniques to large systems, or systems possessing excessive conformational freedom. In this article, the machine learning method kriging is applied to predict both the intra-atomic and interatomic energies, as well as the electrostatic multipole moments, of the atoms of a water molecule at the center of a 10 water molecule (decamer) cluster. Unlike previous work, where the properties of small water clusters were predicted using a molecular local frame, and where training set inputs (features) were based on atomic index, a variety of feature definitions and coordinate frames are considered here to increase prediction accuracy. It is shown that, for a water molecule at the center of a decamer, no single method of defining features or coordinate schemes is optimal for every property. However, explicitly accounting for the structure of the first solvation shell in the definition of the features of the kriging training set, and centring the coordinate frame on the atom-of-interest will, in general, return better predictions than models that apply the standard methods of feature definition, or a molecular coordinate frame. © 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.

© 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.

Keywords: force field design; interacting quantum atoms; kriging; liquid water; machine learning; quantum chemical topology; quantum theory of atoms in molecules

References

  1. J Comput Chem. 2004 Jul 15;25(9):1157-74 - PubMed
  2. Science. 2005 Jul 1;309(5731):75 - PubMed
  3. J Chem Theory Comput. 2005 Nov;1(6):1096-109 - PubMed
  4. J Comput Chem. 2015 Sep 15;36(24):1844-57 - PubMed
  5. J Chem Theory Comput. 2014 Sep 9;10(9):3708-19 - PubMed
  6. Phys Chem Chem Phys. 2009 Aug 14;11(30):6365-76 - PubMed
  7. Phys Rev Lett. 1985 Nov 25;55(22):2471-2474 - PubMed
  8. J Chem Theory Comput. 2015 Jul 14;11(7):3225-33 - PubMed
  9. Phys Chem Chem Phys. 2013 Nov 7;15(41):18249-61 - PubMed
  10. J Chem Theory Comput. 2008 Feb;4(2):353-65 - PubMed
  11. J Comput Chem. 2016 Oct 15;37(27):2409-22 - PubMed
  12. J Mol Graph. 1996 Jun;14(3):136-41 - PubMed
  13. Phys Rev Lett. 2012 Feb 3;108(5):058301 - PubMed
  14. J Chem Theory Comput. 2016 Apr 12;12(4):1499-513 - PubMed
  15. Faraday Discuss. 2009;141:251-76; discussion 309-46 - PubMed
  16. Chemistry. 2013 Oct 11;19(42):14304-15 - PubMed
  17. J Chem Theory Comput. 2009 Jun 9;5(6):1474-89 - PubMed
  18. J Chem Phys. 2010 May 7;132(17):174504 - PubMed
  19. J Comput Chem. 2013 Aug 5;34(21):1850-61 - PubMed

Publication Types