Quantifying configuration-sampling error in Langevin simulations of complex molecular systems

Josh Fass, David Sivak , Gavin E. Crooks, Kyle A. Beauchamp, Benedict Leimkuhler, and John Chodera.
Entropy 20:318, 2018. [DOI] [PDF] [GitHub] [bioRxiv preprint]

Molecular dynamics simulations necessarily use a finite timestep, which introduces error or bias in the sampled configuration space density that grows rapidly with increasing timestep. For the first time, we show how to compute a natural measure of this error---the KL divergence---in both phase and configuration space for a widely used family of Langevin integrators, and show that VRORV is generally superior for simulation of molecular systems.

Approaches for calculating solvation free energies and enthalpies demonstrated with an update of the FreeSolv database

Guilherme Duarte Ramos Matos, Daisy Y. Kyu, Hannes H. Loeffler, John D. Chodera, Michael R. Shirts, David Mobley
Journal of Chemical Engineering Data 62:1559, 2017. [DOI] [bioRxiv] [GitHub]

We review alchemical methods for computing solvation free energies and present an update (version 0.5) to the FreeSolv database of experimental and calculated hydration free energies of neutral compounds.

Ensembler: Enabling high-throughput molecular simulations at the superfamily scale

Daniel L. Parton, Patrick B. Grinaway, Sonya M. Hanson, Kyle A. Beauchamp, and John D. Chodera
PLoS Computational Biology 12:e1004728, 2016. [DOI] [PDF] [bioRxiv] / data: [Dryad] / code: [GitHub]

We demonstrate a new tool that enables---for the first time---massively parallel molecular simulation studies of biomolecular dynamics at the superfamily scale, illustrating its application to protein tyrosine kinases, an important class of drug targets in cancer.

A simple method for automated equilibration detection in molecular simulations

John D. Chodera.
J. Chem. Theor. Comput. 12:1799, 2016. [DOI] [PDF] / code to reproduce manuscript: [GitHub] / preprint: [bioRxiv] / available in pymbar.timeseries

We present a simple scheme for automatically selecting how much initial simulation data to discard to equilibration or burn-in based on maximizing the number of statistically uncorrelated samples in the dataset.

Keywords: molecular simulation; molecular dynamics; burn-in; equilibration; production; analysis

Towards Automated Benchmarking of Atomistic Forcefields: Neat Liquid Densities and Static Dielectric Constants from the ThermoML Data Archive

Kyle A. Beauchamp, Julie M. Behr, Ariën S. Rustenburg, Christopher I. Bayly, Kenneth Kroenlein, and John D. Chodera.
J. Phys. Chem. B 119:12912, 2015. [DOI] [PDF] // code: [GitHub] // preprint: [arXiv] 

Progress in forcefield validation and parameterization has been hindered by the availability of high-quality machine-readable physical property data for small organic molecules. We show how the NIST ThermoML dataset provides a solution to this problem, and demonstrate its utility in benchmarking the GAFF/AM1-BCC small molecule forcefield on neat liquid densities and static dielectric constants to uncover problems in the representation of low-dielectric environments.

Spectral rate theory for two-state kinetics

Jan-Hendrik Prinz, John D. Chodera, and Frank Noé.
Phys. Rev. X 4:011020, 2014. [DOI] [PDF]

We present a new mathematical framework for unifying various two-state rate theories presented in the physical chemistry literature over many decades, and provide a quantitative way to measure reaction coordinate quality.

Identifying ligand binding sites and poses using GPU-accelerated Hamiltonian replica exchange molecular dynamics

Kai Wang K, John D. Chodera, Yanzhi Yang, and Michael R. Shirts. 
J. Comput. Aid. Mol. Des. 27:989, 2013. [DOI] [PDF]

We show how bound ligand poses can be identified even when the location of the binding sites are unknown using the machinery of alchemical modern free energy calculations on graphics processors. 

Systematic improvement of a classical molecular model of water

Lee-Ping Wang, Teresa L. Head-Gordon, Jay W. Ponder, Pengyu Ren, John D. Chodera, Peter K. Eastman, Todd J. Martinez, and Vijay S. Pande.
J. Phys. Chem. B 117:9956, 2013. [DOI] [PDF]

A new inexpensive polarizable model of liquid water for next-generation forcefields is derived using an automated parameterization engine.

Using nonequilibrium fluctuation theorems to understand and correct errors in equilibrium and nonequilibrium discrete Langevin dynamics simulations

David A. Sivak, John D. Chodera, and Gavin E. Crooks.
Phys. Rev. X 3:011007, 2013. [DOI] [PDF]

The finite-timestep errors in molecular dynamics simulations can be interpreted as a form of nonequilibrium work.  We show how this leads to straightforward schemes for correcting for these errors or assessing their impact.

Keywords: velocity verlet with Velocity randomization; VVVR; nonequilibrium free energy; integrator error; nonequilibrium integration

OpenMM 4: A reusable, extensible, hardware independent library for high performance molecular simulation

Peter Eastman, Mark S. Friedrichs, John D. Chodera, Randy J. Radmer, Chris M. Bruns, Joy P. Ku, Kyle A. Beauchamp, T. J. Lane, Lee-Ping Wang, Diwakar Shukla, Tony Tye, Mike Houston, Timo Stich, Christoph Klein, Michael R. Shirts, and Vijay S. Pande.
J. Chem. Theor. Comput. 9:461, 2013. [DOI] [PDF]

We describe the latest version of an open-source, GPU-accelerated library and toolkit for molecular simulation.

Replica exchange and expanded ensemble simulations as Gibbs sampling: Simple improvements for enhanced mixing

John D. Chodera and Michael R. Shirts.
J. Chem. Phys. 135:194110, 2011. [DOI] [PDF] 

We show how a simple change to the way exchanges are handled in the popular replica-exchange simulation methodology can enormously increase efficiency at no increase in computational cost.

Nonequilibrium candidate Monte Carlo is an efficient tool for equilibrium simulation

Jerome P. Nilmeier, Gavin E. Crooks, David D. L. Minh, and John D. Chodera. 
Proc. Natl. Acad. Sci. USA 108:E1009, 2011. [DOI] [PDF]

We present a significant generalization of Monte Carlo methods that provide an enormously useful tool for enhancing the efficiency of molecular simulations and enabling molecular design.

Keywords: NCMC; Monte Carlo; Metropolis-Hastings; acceptance rates; molecular dynamics

The Social Network (of protein conformations)

John D. Chodera and Vijay S. Pande.
Proc. Natl. Acad. Sci. USA 108:12969, 2011. [DOI] [PDF]

Commentary: A new methodology for mapping protein conformational spaces is reminiscent of how we use two-dimensional maps to navigate a three-dimensional world.

Dynamical reweighting: Improved estimates for dynamical properties from simulations at multiple temperatures

John D. Chodera, William C. Swope, Frank Noé, Jan-Hendrik Prinz, Michael R. Shirts, and Vijay S. Pande.
J. Chem. Phys. 134:244107, 2011. [DOI] [PDF] 

We describe how reweighing techniques can provide optimal estimates of temperature-dependent dynamical properties from simulations conducted at multiple temperatures.

Optimal use of data in parallel tempering simulations for the construction of discrete-state Markov models of biomolecular dynamics

Jan-Hendrik Prinz, John D. Chodera, Vijay S. Pande, Jeremy C. Smith, and Frank Noé.
J. Chem. Phys. 134:244108, 2011. [DOI] [PDF]

Markov models of molecular kinetics: Generation and validation

Jan-Hendrik Prinz, Hao Wu, Marco Sarich, Bettina Keller, Martin Fischbach, Martin Held, John D. Chodera, Christof Schüttle, and Frank Noé.
J. Chem. Phys. 134:174105, 2011. [DOI] [PDF]

A review of current best practices for the generation and validation of Markov state models for describing the stochastic dynamics of biomolecular systems.

Probability distributions of molecular observables computed from Markov models. II. Uncertainties in observables and their time-evolution

John D. Chodera and Frank Noé.
J. Chem. Phys. 133:105102, 2010. [DOI] [PDF]

A simple Bayesian approach for the modeling of statistical uncertainties in kinetic and equilibrium quantities computed from Markov state models of biomolecular dynamics.

The mechanical properties of PCNA: Implications for the loading and function of a DNA sliding clamp

Joshua L. Adelman, John D. Chodera, I. W. Kuo, Thomas F. Miller III, and Daniel Barsky.
Biophys. J. 98:3062, 2010. Featured on issue cover. [DOI] [PDF]

Molecular simulations of the PCNA clamp responsible for DNA polymerase processivity show a surprisingly small energetic penalty for the deformation required for clamp loading.

Current status of the AMOEBA polarizable force field

Jay W. Ponder, Chuanjie Wu, Pengyu Ren, Vijay S. Pande, John D. Chodera, David L. Mobley, Michael J. Schnieders, Imran Haque, David S. Lambrecht, Robert A. DiStasio Jr., Martin Head-Gordon, Gary N. I. Clark, Margaret E. Johnson, and Teresa Head-Gordon.
J. Phys. Chem. B 114:2549, 2010. [DOI] [PDF]

A report on the status of the AMOEBA polarizable force field and its ability to reproduce a diverse set of physical chemical phenomenon to high accuracy.

Bayesian comparison of Markov models of molecular dynamics with detailed balance constraint

Sergio Bacallado, John D. Chodera, and Vijay S. Pande.
J. Chem. Phys. 131:045106, 2009. [DOI] [PDF] 

A Bayesian scheme for comparing state space decompositions for Markov state models of biomolecular dynamics that incorporates the fact that physical systems must obey detailed balance.  This paper utilizes recent results from Markov chain theory on edge-reinforced random walks.