skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Machine Learning Force Field Parameters from Ab Initio Data

Abstract

Machine learning (ML) techniques with the genetic algorithm (GA) have been applied to determine a polarizable force field parameters using only ab initio data from quantum mechanics (QM) calculations of molecular clusters at the MP2/6-31G(d,p), DFMP2(fc)/jul-cc-pVDZ, and DFMP2(fc)/jul-cc-pVTZ levels to predict experimental condensed phase properties (i.e., density and heat of vaporization). The performance of this ML/GA approach is demonstrated on 4943 dimer electrostatic potentials and 1250 cluster interaction energies for methanol. Excellent agreement between the training data set from QM calculations and the optimized force field model was achieved. The results were further improved by introducing an offset factor during the machine learning process to compensate for the discrepancy between the QM calculated energy and the energy reproduced by optimized force field, while maintaining the local “shape” of the QM energy surface. Throughout the machine learning process, experimental observables were not involved in the objective function, but were only used for model validation. The best model, optimized from the QM data at the DFMP2(fc)/jul-cc-pVTZ level, appears to perform even better than the original AMOEBA force field (amoeba09.prm), which was optimized empirically to match liquid properties. The present effort shows the possibility of using machine learning techniques to develop descriptive polarizablemore » force field using only QM data. The ML/GA strategy to optimize force fields parameters described here could easily be extended to other molecular systems.« less

Authors:
 [1];  [2]; ORCiD logo [3];  [4];  [4]; ORCiD logo [5];  [5];  [3]; ORCiD logo [6]
  1. Argonne Leadership Computing Facility, Argonne National Laboratory, Argonne, Illinois 60439, United States
  2. Department of Biochemistry and Molecular Biophysics, University of Chicago, Chicago, Illinois 60637, United States
  3. Laboratory of Computational Biology, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892, United States
  4. Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois 60439, United States
  5. Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois 60439, United States; Computational Institute, University of Chicago, Chicago, Illinois 60637, United States
  6. Department of Biochemistry and Molecular Biophysics, University of Chicago, Chicago, Illinois 60637, United States; Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois 60439, United States; Computational Institute, University of Chicago, Chicago, Illinois 60637, United States
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
OSTI Identifier:
1415482
DOE Contract Number:  
AC02-06CH11357
Resource Type:
Journal Article
Resource Relation:
Journal Name: Journal of Chemical Theory and Computation; Journal Volume: 13; Journal Issue: 9
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Force Field; Machine Learning; ab initio Data

Citation Formats

Li, Ying, Li, Hui, Pickard, Frank C., Narayanan, Badri, Sen, Fatih G., Chan, Maria K. Y., Sankaranarayanan, Subramanian K. R. S., Brooks, Bernard R., and Roux, Benoît. Machine Learning Force Field Parameters from Ab Initio Data. United States: N. p., 2017. Web. doi:10.1021/acs.jctc.7b00521.
Li, Ying, Li, Hui, Pickard, Frank C., Narayanan, Badri, Sen, Fatih G., Chan, Maria K. Y., Sankaranarayanan, Subramanian K. R. S., Brooks, Bernard R., & Roux, Benoît. Machine Learning Force Field Parameters from Ab Initio Data. United States. doi:10.1021/acs.jctc.7b00521.
Li, Ying, Li, Hui, Pickard, Frank C., Narayanan, Badri, Sen, Fatih G., Chan, Maria K. Y., Sankaranarayanan, Subramanian K. R. S., Brooks, Bernard R., and Roux, Benoît. Fri . "Machine Learning Force Field Parameters from Ab Initio Data". United States. doi:10.1021/acs.jctc.7b00521.
@article{osti_1415482,
title = {Machine Learning Force Field Parameters from Ab Initio Data},
author = {Li, Ying and Li, Hui and Pickard, Frank C. and Narayanan, Badri and Sen, Fatih G. and Chan, Maria K. Y. and Sankaranarayanan, Subramanian K. R. S. and Brooks, Bernard R. and Roux, Benoît},
abstractNote = {Machine learning (ML) techniques with the genetic algorithm (GA) have been applied to determine a polarizable force field parameters using only ab initio data from quantum mechanics (QM) calculations of molecular clusters at the MP2/6-31G(d,p), DFMP2(fc)/jul-cc-pVDZ, and DFMP2(fc)/jul-cc-pVTZ levels to predict experimental condensed phase properties (i.e., density and heat of vaporization). The performance of this ML/GA approach is demonstrated on 4943 dimer electrostatic potentials and 1250 cluster interaction energies for methanol. Excellent agreement between the training data set from QM calculations and the optimized force field model was achieved. The results were further improved by introducing an offset factor during the machine learning process to compensate for the discrepancy between the QM calculated energy and the energy reproduced by optimized force field, while maintaining the local “shape” of the QM energy surface. Throughout the machine learning process, experimental observables were not involved in the objective function, but were only used for model validation. The best model, optimized from the QM data at the DFMP2(fc)/jul-cc-pVTZ level, appears to perform even better than the original AMOEBA force field (amoeba09.prm), which was optimized empirically to match liquid properties. The present effort shows the possibility of using machine learning techniques to develop descriptive polarizable force field using only QM data. The ML/GA strategy to optimize force fields parameters described here could easily be extended to other molecular systems.},
doi = {10.1021/acs.jctc.7b00521},
journal = {Journal of Chemical Theory and Computation},
number = 9,
volume = 13,
place = {United States},
year = {Fri Aug 11 00:00:00 EDT 2017},
month = {Fri Aug 11 00:00:00 EDT 2017}
}