Machine Learning Force Field Parameters from Ab Initio Data
| Autor: | Ying Li, Badri Narayanan, Bernard R. Brooks, Maria K. Y. Chan, Hui Li, Subramanian K. R. S. Sankaranarayanan, Benoît Roux, Fatih Şen, Frank C. Pickard |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Offset (computer science)
Training set 010304 chemical physics business.industry Chemistry Ab initio Observable 02 engineering and technology Enthalpy of vaporization 021001 nanoscience & nanotechnology Machine learning computer.software_genre 01 natural sciences Article Force field (chemistry) Computer Science Applications Polarizability 0103 physical sciences Cluster (physics) Artificial intelligence Physical and Theoretical Chemistry Atomic physics 0210 nano-technology business computer |
| Zdroj: | Journal of Chemical Theory and Computation. 13:4492-4503 |
| ISSN: | 1549-9626 1549-9618 |
| DOI: | 10.1021/acs.jctc.7b00521 |
| Popis: | Machine learning (ML) techniques with the genetic algorithm (GA) have been applied to explore 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 4,943 dimers electrostatic potentials and 1,250 clusters interaction energies for methanol. Excellent agreement between the training dataset from QM calculations and the optimized force field model can be achieved. Better results are achieved by introducing an offset factor during the machine learning process to compensate for the discrepancy of the QM calculated energy and the energy reproduced by optimized force field, where the offset factor maintain 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 field parameters described here could easily be extended to other molecular systems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|