Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics

Tsourtis, Anastasios; Pantazis, Yannis; Harmandaris, Vagelis

doi:10.1063/1.4922924

Title: Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics

Full Record
Other Related Research

Abstract

In this paper, we present a parametric sensitivity analysis (SA) methodology for continuous time and continuous space Markov processes represented by stochastic differential equations. Particularly, we focus on stochastic molecular dynamics as described by the Langevin equation. The utilized SA method is based on the computation of the information-theoretic (and thermodynamic) quantity of relative entropy rate (RER) and the associated Fisher information matrix (FIM) between path distributions, and it is an extension of the work proposed by Y. Pantazis and M. A. Katsoulakis [J. Chem. Phys. 138, 054115 (2013)]. A major advantage of the pathwise SA method is that both RER and pathwise FIM depend only on averages of the force field; therefore, they are tractable and computable as ergodic averages from a single run of the molecular dynamics simulation both in equilibrium and in non-equilibrium steady state regimes. We validate the performance of the extended SA method to two different molecular stochastic systems, a standard Lennard-Jones fluid and an all-atom methane liquid, and compare the obtained parameter sensitivities with parameter sensitivities on three popular and well-studied observable functions, namely, the radial distribution function, the mean squared displacement, and the pressure. Results show that the RER-based sensitivities are highly correlatedmore »« less

Authors:

Tsourtis, Anastasios ^[1]; Pantazis, Yannis; Harmandaris, Vagelis ^[2]

Department of Mathematics and Applied Mathematics, University of Crete, Crete (Greece)
Department of Mathematics and Applied Mathematics, University of Crete, and Institute of Applied and Computational Mathematics (IACM), Foundation for Research and Technology Hellas (FORTH), GR-70013 Heraklion, Crete (Greece)

Publication Date:: Tue Jul 07 00:00:00 EDT 2015

OSTI Identifier:: 22489546

Resource Type:: Journal Article

Journal Name:: Journal of Chemical Physics

Additional Journal Information:: Journal Volume: 143; Journal Issue: 1; Other Information: (c) 2015 AIP Publishing LLC; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0021-9606

Country of Publication:: United States

Language:: English

Subject:: 37 INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL CHEMISTRY; 97 MATHEMATICAL METHODS AND COMPUTING; COMPARATIVE EVALUATIONS; DIFFERENTIAL EQUATIONS; ENTROPY; EQUILIBRIUM; LANGEVIN EQUATION; LIQUIDS; MARKOV PROCESS; METHANE; MOLECULAR DYNAMICS METHOD; SENSITIVITY ANALYSIS; SIMULATION; SPATIAL DISTRIBUTION

Citation Formats


                    Tsourtis, Anastasios, Pantazis, Yannis, Katsoulakis, Markos A., E-mail: markos@math.umass.edu, and Harmandaris, Vagelis. Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics.  United States: N. p., 2015. 
        Web.  doi:10.1063/1.4922924.

Copy to clipboard


                    Tsourtis, Anastasios, Pantazis, Yannis, Katsoulakis, Markos A., E-mail: markos@math.umass.edu, & Harmandaris, Vagelis. Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics.  United States.  https://doi.org/10.1063/1.4922924

Copy to clipboard


                    Tsourtis, Anastasios, Pantazis, Yannis, Katsoulakis, Markos A., E-mail: markos@math.umass.edu, and Harmandaris, Vagelis. 2015.  
        "Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics".  United States.  https://doi.org/10.1063/1.4922924.

Copy to clipboard


                    
@article{osti_22489546,

  title        = {Parametric sensitivity analysis for stochastic molecular systems using information theoretic metrics},

  author       = {Tsourtis, Anastasios and Pantazis, Yannis and Katsoulakis, Markos A., E-mail: markos@math.umass.edu and Harmandaris, Vagelis},

  abstractNote = {In this paper, we present a parametric sensitivity analysis (SA) methodology for continuous time and continuous space Markov processes represented by stochastic differential equations. Particularly, we focus on stochastic molecular dynamics as described by the Langevin equation. The utilized SA method is based on the computation of the information-theoretic (and thermodynamic) quantity of relative entropy rate (RER) and the associated Fisher information matrix (FIM) between path distributions, and it is an extension of the work proposed by Y. Pantazis and M. A. Katsoulakis [J. Chem. Phys. 138, 054115 (2013)]. A major advantage of the pathwise SA method is that both RER and pathwise FIM depend only on averages of the force field; therefore, they are tractable and computable as ergodic averages from a single run of the molecular dynamics simulation both in equilibrium and in non-equilibrium steady state regimes. We validate the performance of the extended SA method to two different molecular stochastic systems, a standard Lennard-Jones fluid and an all-atom methane liquid, and compare the obtained parameter sensitivities with parameter sensitivities on three popular and well-studied observable functions, namely, the radial distribution function, the mean squared displacement, and the pressure. Results show that the RER-based sensitivities are highly correlated with the observable-based sensitivities.},

  doi          = {10.1063/1.4922924},

  url          = {https://www.osti.gov/biblio/22489546},
  journal      = {Journal of Chemical Physics},

  issn         = {0021-9606},

  number       = 1,

  volume       = 143,

  place        = {United States},

  year         = {Tue Jul 07 00:00:00 EDT 2015},

  month        = {Tue Jul 07 00:00:00 EDT 2015}

}

Copy to clipboard

Journal Article:

https://doi.org/10.1063/1.4922924

Other availability

Find in Google Scholar

Search WorldCat to find libraries that may hold this journal

Save / Share:

Export Metadata

Save to My Library

Similar records in OSTI.GOV collections:

ISAP-MATLAB Package for Sensitivity Analysis of High-Dimensional Stochastic Chemical Networks

Journal Article Hu, Weilong; Pantazis, Yannis; Katsoulakis, Markos - Journal of Statistical Software

Stochastic simulation and modeling play an important role to elucidate the fundamental mechanisms in complex biochemical networks. The parametric sensitivity analysis of reaction networks becomes a powerful mathematical and computational tool, yielding information regarding the robustness and the identifiability of model parameters. However, due to overwhelming computational cost, parametric sensitivity analysis is a extremely challenging problem for stochastic models with a high-dimensional parameter space and for which existing approaches are very slow. Here we present an information-theoretic sensitivity analysis in path-space (ISAP) MATLAB package that simulates stochastic processes with various algorithms and most importantly implements a gradient-free approach to quantifymore »« less
Cited by 1
https://doi.org/10.18637/jss.v085.i03

Full Text Available
Generalized Brownian dynamics. III. Vibrational relaxation of heteronuclear diatomics in rare gases

Journal Article Zhang, J; Harris, C - Journal of Chemical Physics; (USA)

A simple, classical stochastic model has been developed to study the vibrational energy relaxation of heteronuclear diatomic molecules in simple monatomic liquids. This work is an extension of an earlier study on homonuclear diatomic molecules (D. E. Smith and C. B. Harris, J. Chem. Phys. {bold 92}, 1312 (1990)). The model is based on the generalized Langevin equation for generalized Brownian dynamics. The memory function of each atom in the diatomic is determined directly from classical molecular dynamics simulations in the solvent of interest and then scaled by a screening factor. The memory function is modeled using an autoregressive technique,more »« less
https://doi.org/10.1063/1.460494
Generalized Brownian dynamics. II. Vibrational relaxation of diatomic molecules in solution

Journal Article Smith, D; Harris, C - Journal of Chemical Physics; (USA)

A simple classical stochastic model for diatomic vibrational relaxation, based on the generalized Langevin equation, is presented. The memory function in the generalized Langevin equation is determined directly from equilibrium force autocorrelation functions for the individual atoms of the diatomic dissolved in the solvent of interest. A simple autoregressive (AR) procedure, developed in a preceding paper (D. E. Smith and C. B. Harris, J. Chem. Phys. {bold 92}, 1304 (1990)), is used for modeling the memory functions to arbitrary order. This model is tested on the system of iodine in Lennard--Jones xenon using fourth order AR approximations for the memorymore »« less
https://doi.org/10.1063/1.458141
First Principles Simulations fo the Supercritical Behavior of Ore Forming Fluids

Technical Report Weare, John

Abstract of Selected Research Progress: I. First-principles simulation of solvation structure and deprotonation reactions of ore forming metal ions in very nonideal solutions: Advances in algorithms and computational performance achieved in this grant period have allowed the atomic level dynamical simulation of complex nanoscale materials using interparticle forces calculated directly from an accurate density functional solution to the electronic Schr dinger equation (ab-initio molecular dynamics, AIMD). Focus of this program was on the prediction and analysis of the properties of environmentally important ions in aqueous solutions. AIMD methods have provided chemical interpretations of these very complex systems with an unprecedentedmore »« less
https://doi.org/10.2172/1074367

Full Text Available
Mathematical Foundations for Uncertainty Quantification in Materials Design. Final Report

Technical Report Katsoulakis, Markos; Rey-Bellet, Luc

Final report DE-SC0010723. Our key accomplishment in the first year of the grant is that we started developing pathwise information theory-based and goal-oriented sensitivity analysis and parameter identification methods for complex high-dimensional dynamics and in particular of nonequilibrium extended systems. The combination of these novel methodologies provide the first methods in the literature which are capable to handle UQ questions for stochastic complex systems with some or all of the following features: (a) multi-scale models with a very large number of parameters, (b) spatially distributed systems such as Kinetic Monte Carlo or Langevin Dynamics, (c) non-equilibrium processes typically associated withmore »« less
https://doi.org/10.2172/1483471

Full Text Available

Similar Records