Learning free energy landscapes using artificial neural networks.

Existing adaptive bias techniques, which seek to estimate free energies and physical properties from molecular simulations, are limited by their reliance on fixed kernels or basis sets which hinder their ability to efficiently conform to varied free energy landscapes. Further, user-specified parameters are in general non-intuitive yet significantly affect the convergence rate and accuracy of the free energy estimate. Here we propose a novel method, wherein artificial neural networks (ANNs) are used to develop an adaptive biasing potential which learns free energy landscapes. We demonstrate that this method is capable of rapidly adapting to complex free energy landscapes and is not prone to boundary or oscillation problems. The method is made robust to hyperparameters and overfitting through Bayesian regularization which penalizes network weights and auto-regulates the number of effective parameters in the network. ANN sampling represents a promising innovative approach which can resolve complex free energy landscapes in less time than conventional approaches while requiring minimal user input.

[1]  Haohao Fu,et al.  Extended Adaptive Biasing Force Algorithm. An On-the-Fly Implementation for Accurate Free-Energy Calculations. , 2016, Journal of chemical theory and computation.

[2]  Konstantina S. Nikita,et al.  A computer-aided diagnostic system to characterize CT focal liver lesions: design and optimization of a neural network classifier , 2003, IEEE Transactions on Information Technology in Biomedicine.

[3]  Hsiao-Tien Pao,et al.  Forecasting electricity market pricing using artificial neural networks , 2007 .

[4]  M. Mezei.  Adaptive umbrella sampling: Self-consistent determination of the non-Boltzmann bias , 1987 .

[5]  A. Laio,et al.  Escaping free-energy minima , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[6]  N. Abbott,et al.  Measuring liquid crystal elastic constants with free energy perturbations. , 2014, Soft matter.

[7]  Martin T. Hagan,et al.  Neural network design , 1995 .

[8]  Massimiliano Bonomi,et al.  Free-energy landscape of protein oligomerization from atomistic simulations , 2013, Proceedings of the National Academy of Sciences.

[9]  R. Hooft,et al.  An adaptive umbrella sampling procedure in conformational analysis using molecular dynamics and its application to glycol , 1992 .

[10]  G. Torrie,et al.  Nonphysical sampling distributions in Monte Carlo free-energy estimation: Umbrella sampling , 1977 .

[11]  Michele Parrinello,et al.  Generalized neural-network representation of high-dimensional potential-energy surfaces. , 2007, Physical review letters.

[12]  Michele Parrinello,et al.  Variational approach to enhanced sampling and free energy calculations. , 2014, Physical review letters.

[13]  M Karplus,et al.  Evolutionary optimization in quantitative structure-activity relationship: an application of genetic neural networks. , 1996, Journal of medicinal chemistry.

[14]  M. Parrinello,et al.  Well-tempered metadynamics: a smoothly converging and tunable free-energy method. , 2008, Physical review letters.

[15]  David J. C. MacKay,et al.  A Practical Bayesian Framework for Backpropagation Networks , 1992, Neural Computation.

[16]  Juan J. de Pablo,et al.  Flux Tempered Metadynamics , 2011 .

[17]  P. Wolynes,et al.  Topology, structures, and energy landscapes of human chromosomes , 2015, Proceedings of the National Academy of Sciences.

[18]  Christian Bartels,et al.  Multidimensional adaptive umbrella sampling: Applications to main chain and side chain peptide conformations , 1997 .

[19]  Aaron R Dinner,et al.  Automatic method for identifying reaction coordinates in complex systems. , 2005, The journal of physical chemistry. B.

[20]  S. Snyder,et al.  Proceedings of the National Academy of Sciences , 1999 .

[21]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.

[22]  Michele Parrinello,et al.  Enhancing Important Fluctuations: Rare Events and Metadynamics from a Conceptual Viewpoint. , 2016, Annual review of physical chemistry.

[23]  Christophe Chipot,et al.  The Adaptive Biasing Force Method: Everything You Always Wanted To Know but Were Afraid To Ask , 2014, The journal of physical chemistry. B.

[24]  R. Swendsen,et al.  THE weighted histogram analysis method for free‐energy calculations on biomolecules. I. The method , 1992 .

[25]  Stephen J. Wright,et al.  Numerical Optimization (Springer Series in Operations Research and Financial Engineering) , 2000 .

[26]  Juan J de Pablo,et al.  Basis function sampling: a new paradigm for material property computation. , 2014, Physical review letters.

[27]  Juan de Pablo,et al.  A boundary correction algorithm for metadynamics in multiple dimensions. , 2013, The Journal of chemical physics.

[28]  J. Herskowitz,et al.  Proceedings of the National Academy of Sciences, USA , 1996, Current Biology.

[29]  Grant M. Rotskoff,et al.  Transition-Tempered Metadynamics: Robust, Convergent Metadynamics via On-the-Fly Transition Barrier Estimation. , 2014, Journal of chemical theory and computation.

[30]  Noam Bernstein,et al.  Free Energy Surface Reconstruction from Umbrella Samples Using Gaussian Process Regression. , 2013, Journal of chemical theory and computation.

[31]  Rustam Z. Khaliullin,et al.  Graphite-diamond phase coexistence study employing a neural-network mapping of the ab initio potential energy surface , 2010 .

[32]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[33]  Mark E Tuckerman,et al.  Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces. , 2017, Physical review letters.

[34]  W. Cheney,et al.  Numerical analysis: mathematics of scientific computing (2nd ed) , 1991 .

[35]  Zhenwei Li,et al.  Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces. , 2015, Physical review letters.

[36]  Juan J de Pablo,et al.  Sculpting bespoke mountains: Determining free energies with basis expansions. , 2015, The Journal of chemical physics.

[37]  Yuji Sugita,et al.  Neural Network and Nearest Neighbor Algorithms for Enhancing Sampling of Molecular Dynamics. , 2017, Journal of chemical theory and computation.

[38]  Eric F Darve,et al.  Calculating free energies using average force , 2001 .

[39]  Peter G Wolynes,et al.  Simulation studies of the fidelity of biomolecular structure ensemble recreation. , 2006, The Journal of chemical physics.

[40]  Martin Karplus,et al.  Probability Distributions for Complex Systems: Adaptive Umbrella Sampling of the Potential Energy , 1998 .

[41]  M. Parrinello,et al.  Metadynamics with Adaptive Gaussians. , 2012, Journal of chemical theory and computation.

[42]  D. Feng,et al.  IEEE transactions on information technology in biomedicine: special issue on advances in clinical and health-care knowledge management , 2005 .

[43]  W. Marsden.  I and J , 2012 .

[44]  C. Wilmer,et al.  Large-scale screening of hypothetical metal-organic frameworks. , 2012, Nature chemistry.