-
BAD-NEUS: Rapidly converging trajectory stratification
Authors:
John Strahan,
Chatipat Lorpaiboon,
Jonathan Weare,
Aaron R. Dinner
Abstract:
An issue for molecular dynamics simulations is that events of interest often involve timescales that are much longer than the simulation time step, which is set by the fastest timescales of the model. Because of this timescale separation, direct simulation of many events is prohibitively computationally costly. This issue can be overcome by aggregating information from many relatively short simula…
▽ More
An issue for molecular dynamics simulations is that events of interest often involve timescales that are much longer than the simulation time step, which is set by the fastest timescales of the model. Because of this timescale separation, direct simulation of many events is prohibitively computationally costly. This issue can be overcome by aggregating information from many relatively short simulations that sample segments of trajectories involving events of interest. This is the strategy of Markov state models (MSMs) and related approaches, but such methods suffer from approximation error because the variables defining the states generally do not capture the dynamics fully. By contrast, once converged, the weighted ensemble (WE) method aggregates information from trajectory segments so as to yield unbiased estimates of both thermodynamic and kinetic statistics. Unfortunately, errors decay no faster than unbiased simulation in WE. Here we introduce a theoretical framework for describing WE that shows that introduction of an element of stratification, as in nonequilibrium umbrella sampling (NEUS), accelerates convergence. Then, building on ideas from MSMs and related methods, we propose an improved stratification that allows approximation error to be reduced systematically. We show that the improved stratification can decrease simulation times required to achieve a desired precision by orders of magnitude.
△ Less
Submitted 30 April, 2024;
originally announced April 2024.
-
Accurate estimates of dynamical statistics using memory
Authors:
Chatipat Lorpaiboon,
Spencer C. Guo,
John Strahan,
Jonathan Weare,
Aaron R. Dinner
Abstract:
Many chemical reactions and molecular processes occur on timescales that are significantly longer than those accessible by direct simulation. One successful approach to estimating dynamical statistics for such processes is to use many short time series observations of the system to construct a Markov state model (MSM), which approximates the dynamics of the system as memoryless transitions between…
▽ More
Many chemical reactions and molecular processes occur on timescales that are significantly longer than those accessible by direct simulation. One successful approach to estimating dynamical statistics for such processes is to use many short time series observations of the system to construct a Markov state model (MSM), which approximates the dynamics of the system as memoryless transitions between a set of discrete states. The dynamical Galerkin approximation (DGA) generalizes MSMs for the problem of calculating dynamical statistics, such as committors and mean first passage times, by replacing the set of discrete states with a projection onto a basis. Because the projected dynamics are generally not memoryless, the Markov approximation can result in significant systematic error. Inspired by quasi-Markov state models, which employ the generalized master equation to encode memory resulting from the projection, we reformulate DGA to account for memory and analyze its performance on two systems: a two-dimensional triple well and helix-to-helix transitions of the AIB$_9$ peptide. We demonstrate that our method is robust to the choice of basis and can decrease the time series length required to obtain accurate kinetics by an order of magnitude.
△ Less
Submitted 13 November, 2023;
originally announced November 2023.
-
Inexact iterative numerical linear algebra for neural network-based spectral estimation and rare-event prediction
Authors:
John Strahan,
Spencer C. Guo,
Chatipat Lorpaiboon,
Aaron R. Dinner,
Jonathan Weare
Abstract:
Understanding dynamics in complex systems is challenging because there are many degrees of freedom, and those that are most important for describing events of interest are often not obvious. The leading eigenfunctions of the transition operator are useful for visualization, and they can provide an efficient basis for computing statistics such as the likelihood and average time of events (predictio…
▽ More
Understanding dynamics in complex systems is challenging because there are many degrees of freedom, and those that are most important for describing events of interest are often not obvious. The leading eigenfunctions of the transition operator are useful for visualization, and they can provide an efficient basis for computing statistics such as the likelihood and average time of events (predictions). Here we develop inexact iterative linear algebra methods for computing these eigenfunctions (spectral estimation) and making predictions from a data set of short trajectories sampled at finite intervals. We demonstrate the methods on a low-dimensional model that facilitates visualization and a high-dimensional model of a biomolecular system. Implications for the prediction problem in reinforcement learning are discussed.
△ Less
Submitted 20 July, 2023; v1 submitted 22 March, 2023;
originally announced March 2023.
-
Predicting rare events using neural networks and short-trajectory data
Authors:
John Strahan,
Justin Finkel,
Aaron R. Dinner,
Jonathan Weare
Abstract:
Estimating the likelihood, timing, and nature of events is a major goal of modeling stochastic dynamical systems. When the event is rare in comparison with the timescales of simulation and/or measurement needed to resolve the elemental dynamics, accurate prediction from direct observations becomes challenging. In such cases a more effective approach is to cast statistics of interest as solutions t…
▽ More
Estimating the likelihood, timing, and nature of events is a major goal of modeling stochastic dynamical systems. When the event is rare in comparison with the timescales of simulation and/or measurement needed to resolve the elemental dynamics, accurate prediction from direct observations becomes challenging. In such cases a more effective approach is to cast statistics of interest as solutions to Feynman-Kac equations (partial differential equations). Here, we develop an approach to solve Feynman-Kac equations by training neural networks on short-trajectory data. Our approach is based on a Markov approximation but otherwise avoids assumptions about the underlying model and dynamics. This makes it applicable to treating complex computational models and observational data. We illustrate the advantages of our method using a low-dimensional model that facilitates visualization, and this analysis motivates an adaptive sampling strategy that allows on-the-fly identification of and addition of data to regions important for predicting the statistics of interest. Finally, we demonstrate that we can compute accurate statistics for a 75-dimensional model of sudden stratospheric warming. This system provides a stringent test bed for our method.
△ Less
Submitted 2 March, 2023; v1 submitted 2 August, 2022;
originally announced August 2022.
-
Long-timescale predictions from short-trajectory data: A benchmark analysis of the trp-cage miniprotein
Authors:
John Strahan,
Adam Antoszewski,
Chatipat Lorpaiboon,
Bodhi P. Vani,
Jonathan Weare,
Aaron R. Dinner
Abstract:
Elucidating physical mechanisms with statistical confidence from molecular dynamics simulations can be challenging owing to the many degrees of freedom that contribute to collective motions. To address this issue, we recently introduced a dynamical Galerkin approximation (DGA) [Thiede et al. J. Phys. Chem. 150, 244111 (2019)], in which chemical kinetic statistics that satisfy equations of dynamica…
▽ More
Elucidating physical mechanisms with statistical confidence from molecular dynamics simulations can be challenging owing to the many degrees of freedom that contribute to collective motions. To address this issue, we recently introduced a dynamical Galerkin approximation (DGA) [Thiede et al. J. Phys. Chem. 150, 244111 (2019)], in which chemical kinetic statistics that satisfy equations of dynamical operators are represented by a basis expansion. Here, we reformulate this approach, clarifying (and reducing) the dependence on the choice of lag time. We present a new projection of the reactive current onto collective variables and provide improved estimators for rates and committors. We also present simple procedures for constructing suitable smoothly varying basis functions from arbitrary molecular features. To evaluate estimators and basis sets numerically, we generate and carefully validate a dataset of short trajectories for the unfolding and folding of the trp-cage miniprotein, a well-studied system. Our analysis demonstrates a comprehensive strategy for characterizing reaction pathways quantitatively.
△ Less
Submitted 8 September, 2020;
originally announced September 2020.
