Reinforced dynamics for enhanced sampling in large atomic and molecular systems
Abstract
A new approach for efficiently exploring the configuration space and computing the free energy of large atomic and molecular systems is proposed, motivated by an analogy with reinforcement learning. There are two major components in this new approach. Like metadynamics, it allows for an efficient exploration of the configuration space by adding an adaptively computed biasing potential to the original dynamics. Like deep reinforcement learning, this biasing potential is trained on the fly using deep neural networks, with data collected judiciously from the exploration and an uncertainty indicator from the neural network model playing the role of the reward function. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. Finally, the method is illustrated by studying the full-atom explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.
- Authors:
-
- Princeton Univ., Princeton, NJ (United States). Program in Applied and Computational Mathematics
- Inst. of Applied Physics and Computational Mathematics, Beijing (China); CAEP Software Center for High Performance Numerical Simulation, Beijing (China)
- Princeton Univ., Princeton, NJ (United States). Dept. of Mathematics and Program in Applied and Computational Mathematics; Beijing Inst. of Big Data Research, Beijing (China)
- Publication Date:
- Research Org.:
- Princeton Univ., NJ (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC)
- OSTI Identifier:
- 1540165
- Alternate Identifier(s):
- OSTI ID: 1429965
- Grant/Contract Number:
- SC0008626; SC0009248
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Journal of Chemical Physics
- Additional Journal Information:
- Journal Volume: 148; Journal Issue: 12; Journal ID: ISSN 0021-9606
- Publisher:
- American Institute of Physics (AIP)
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; chemistry; physics
Citation Formats
Zhang, Linfeng, Wang, Han, and E, Weinan. Reinforced dynamics for enhanced sampling in large atomic and molecular systems. United States: N. p., 2018.
Web. doi:10.1063/1.5019675.
Zhang, Linfeng, Wang, Han, & E, Weinan. Reinforced dynamics for enhanced sampling in large atomic and molecular systems. United States. https://doi.org/10.1063/1.5019675
Zhang, Linfeng, Wang, Han, and E, Weinan. Tue .
"Reinforced dynamics for enhanced sampling in large atomic and molecular systems". United States. https://doi.org/10.1063/1.5019675. https://www.osti.gov/servlets/purl/1540165.
@article{osti_1540165,
title = {Reinforced dynamics for enhanced sampling in large atomic and molecular systems},
author = {Zhang, Linfeng and Wang, Han and E, Weinan},
abstractNote = {A new approach for efficiently exploring the configuration space and computing the free energy of large atomic and molecular systems is proposed, motivated by an analogy with reinforcement learning. There are two major components in this new approach. Like metadynamics, it allows for an efficient exploration of the configuration space by adding an adaptively computed biasing potential to the original dynamics. Like deep reinforcement learning, this biasing potential is trained on the fly using deep neural networks, with data collected judiciously from the exploration and an uncertainty indicator from the neural network model playing the role of the reward function. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. Finally, the method is illustrated by studying the full-atom explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.},
doi = {10.1063/1.5019675},
journal = {Journal of Chemical Physics},
number = 12,
volume = 148,
place = {United States},
year = {Tue Mar 27 00:00:00 EDT 2018},
month = {Tue Mar 27 00:00:00 EDT 2018}
}
Web of Science
Works referenced in this record:
Spontaneous drying of non-polar deep-cavity cavitand pockets in aqueous solution
journal, May 2020
- Barnett, J. Wesley; Sullivan, Matthew R.; Long, Joshua A.
- Nature Chemistry, Vol. 12, Issue 7
Deep learning
journal, May 2015
- LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
- Nature, Vol. 521, Issue 7553
Multidimensional free-energy calculations using the weighted histogram analysis method
journal, November 1995
- Kumar, Shankar; Rosenberg, John M.; Bouzida, Djamal
- Journal of Computational Chemistry, Vol. 16, Issue 11
Exploration, Sampling, And Reconstruction of Free Energy Surfaces with Gaussian Process Regression.
text, January 2016
- Mones, Letif; Bernstein, Noam; Csányi, Gábor
- Apollo - University of Cambridge Repository
Comparison of simple potential functions for simulating liquid water
journal, July 1983
- Jorgensen, William L.; Chandrasekhar, Jayaraman; Madura, Jeffry D.
- The Journal of Chemical Physics, Vol. 79, Issue 2
GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers
journal, September 2015
- Abraham, Mark James; Murtola, Teemu; Schulz, Roland
- SoftwareX, Vol. 1-2
A temperature accelerated method for sampling free energy and determining reaction pathways in rare events simulations
journal, July 2006
- Maragliano, Luca; Vanden-Eijnden, Eric
- Chemical Physics Letters, Vol. 426, Issue 1-3
Self-adaptive enhanced sampling in the energy and trajectory spaces: Accelerated thermodynamics and kinetic calculations
journal, April 2008
- Gao, Yi Qin
- The Journal of Chemical Physics, Vol. 128, Issue 13
Comparison of multiple Amber force fields and development of improved protein backbone parameters
journal, November 2006
- Hornak, Viktor; Abel, Robert; Okur, Asim
- Proteins: Structure, Function, and Bioinformatics, Vol. 65, Issue 3
PLUMED 2: New feathers for an old bird
journal, February 2014
- Tribello, Gareth A.; Bonomi, Massimiliano; Branduardi, Davide
- Computer Physics Communications, Vol. 185, Issue 2
Efficient Estimation of Rare-Event Kinetics
journal, January 2016
- Trendelkamp-Schroer, Benjamin; Noé, Frank
- Physical Review X, Vol. 6, Issue 1
Deep Potential: A General Representation of a Many-Body Potential Energy Surface
journal, January 2018
- Han, Jiequn; Zhang, Linfeng; Car, Roberto
- Communications in Computational Physics, Vol. 23, Issue 3
Advances and Continuing Challenges in Achieving Realistic and Predictive Simulations of the Properties of Organic and Biological Molecules
journal, January 1996
- Kollman, Peter A.
- Accounts of Chemical Research, Vol. 29, Issue 10
Canonical sampling through velocity rescaling
journal, January 2007
- Bussi, Giovanni; Donadio, Davide; Parrinello, Michele
- The Journal of Chemical Physics, Vol. 126, Issue 1
Efficient estimation of rare-event kinetics
text, January 2014
- Trendelkamp-Schroer, Benjamin; Noe, Frank
- arXiv
Single-sweep methods for free energy calculations
journal, May 2008
- Maragliano, Luca; Vanden-Eijnden, Eric
- The Journal of Chemical Physics, Vol. 128, Issue 18
Blue Moon Sampling, Vectorial Reaction Coordinates, and Unbiased Constrained Dynamics
journal, September 2005
- Ciccotti, Giovanni; Kapral, Raymond; Vanden-Eijnden, Eric
- ChemPhysChem, Vol. 6, Issue 9
Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method
text, January 2008
- Barducci, Alessandro; Bussi, Giovanni; Parrinello, Michele
- arXiv
Multiple free energies from a single simulation: Extending enveloping distribution sampling to nonoverlapping phase-space distributions
journal, May 2008
- Christ, Clara D.; van Gunsteren, Wilfred F.
- The Journal of Chemical Physics, Vol. 128, Issue 17
THE weighted histogram analysis method for free-energy calculations on biomolecules. I. The method
journal, October 1992
- Kumar, Shankar; Rosenberg, John M.; Bouzida, Djamal
- Journal of Computational Chemistry, Vol. 13, Issue 8
Single-Sweep Methods for Free Energy Calculations
text, January 2007
- Maragliano, Luca; Vanden-Eijnden, Eric
- arXiv
The “Hot-Solvent/Cold-Solute” Problem Revisited
journal, July 2008
- Lingenheil, M.; Denschlag, R.; Reichold, R.
- Journal of Chemical Theory and Computation, Vol. 4, Issue 8
Free Energy Surface Reconstruction from Umbrella Samples Using Gaussian Process Regression
journal, August 2014
- Stecher, Thomas; Bernstein, Noam; Csányi, Gábor
- Journal of Chemical Theory and Computation, Vol. 10, Issue 9
Rational identification and characterisation of peptide ligands for targeting polysialic acid
journal, May 2020
- Shastry, Divya G.; Irudayanathan, Flaviyan Jerome; Williams, Asher
- Scientific Reports, Vol. 10, Issue 1
Polymorphic transitions in single crystals: A new molecular dynamics method
journal, December 1981
- Parrinello, M.; Rahman, A.
- Journal of Applied Physics, Vol. 52, Issue 12
Hyperdynamics: Accelerated Molecular Dynamics of Infrequent Events
journal, May 1997
- Voter, Arthur F.
- Physical Review Letters, Vol. 78, Issue 20
Simplified and improved string method for computing the minimum energy paths in barrier-crossing events
journal, April 2007
- E., Weinan; Ren, Weiqing; Vanden-Eijnden, Eric
- The Journal of Chemical Physics, Vol. 126, Issue 16
Deep Potential: a general representation of a many-body potential energy surface
text, January 2017
- Han, Jiequn; Zhang, Linfeng; Car, Roberto
- arXiv
Efficient and Direct Generation of Multidimensional Free Energy Surfaces via Adiabatic Dynamics without Coordinate Transformations
journal, December 2008
- Abrams, Jerry B.; Tuckerman, Mark E.
- The Journal of Physical Chemistry B, Vol. 112, Issue 49
Settle: An analytical version of the SHAKE and RATTLE algorithm for rigid water models
journal, October 1992
- Miyamoto, Shuichi; Kollman, Peter A.
- Journal of Computational Chemistry, Vol. 13, Issue 8
String Method for the Study of Rare Events
text, January 2002
- E., Weinan; Ren, Weiqing; Vanden-Eijnden, Eric
- arXiv
Deep Potential Molecular Dynamics: A Scalable Model with the Accuracy of Quantum Mechanics
journal, April 2018
- Zhang, Linfeng; Han, Jiequn; Wang, Han
- Physical Review Letters, Vol. 120, Issue 14
Replica-exchange molecular dynamics method for protein folding
journal, November 1999
- Sugita, Yuji; Okamoto, Yuko
- Chemical Physics Letters, Vol. 314, Issue 1-2
Efficient Estimation of Rare-Event Kinetics
journal, January 2016
- Trendelkamp-Schroer, Benjamin; Noé, Frank
- Physical Review X, Vol. 6, Issue 1
A smooth particle mesh Ewald method
journal, November 1995
- Essmann, Ulrich; Perera, Lalith; Berkowitz, Max L.
- The Journal of Chemical Physics, Vol. 103, Issue 19
Multiple free energies from a single simulation: Extending enveloping distribution sampling to nonoverlapping phase-space distributions
journal, May 2008
- Christ, Clara D.; van Gunsteren, Wilfred F.
- The Journal of Chemical Physics, Vol. 128, Issue 17
Escaping free-energy minima
journal, September 2002
- Laio, A.; Parrinello, M.
- Proceedings of the National Academy of Sciences, Vol. 99, Issue 20
Exploration, Sampling, And Reconstruction of Free Energy Surfaces with Gaussian Process Regression
journal, September 2016
- Mones, Letif; Bernstein, Noam; Csányi, Gábor
- Journal of Chemical Theory and Computation, Vol. 12, Issue 10
Efficient multidimensional free energy calculations for ab initio molecular dynamics using classical bias potentials
journal, January 2000
- VandeVondele, Joost; Rothlisberger, Ursula
- The Journal of Chemical Physics, Vol. 113, Issue 12
Neural Network and Nearest Neighbor Algorithms for Enhancing Sampling of Molecular Dynamics
journal, May 2017
- Galvelis, Raimondas; Sugita, Yuji
- Journal of Chemical Theory and Computation, Vol. 13, Issue 6
Works referencing / citing this record:
DeePCG: Constructing coarse-grained models via deep neural networks
journal, July 2018
- Zhang, Linfeng; Han, Jiequn; Wang, Han
- The Journal of Chemical Physics, Vol. 149, Issue 3
Deep learning inter-atomic potential model for accurate irradiation damage simulations
text, January 2019
- Wang, Hao; Guo, Xun; Zhang, Linfeng
- arXiv
DeePCG: constructing coarse-grained models via deep neural networks
text, January 2018
- Zhang, Linfeng; Han, Jiequn; Wang, Han
- arXiv
Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
text, January 2018
- Zhang, Linfeng; Lin, De-Ye; Wang, Han
- arXiv
DeePCG: constructing coarse-grained models via deep neural networks
text, January 2018
- Zhang, Linfeng; Han, Jiequn; Wang, Han
- arXiv
Deep learning inter-atomic potential model for accurate irradiation damage simulations
journal, June 2019
- Wang, Hao; Guo, Xun; Zhang, Linfeng
- Applied Physics Letters, Vol. 114, Issue 24