Calibration of Boltzmann distribution priors in Bayesian data analysis.

The Boltzmann distribution is commonly used as a prior probability in Bayesian data analysis. Examples include the Ising model in statistical image analysis and the canonical ensemble based on molecular dynamics force fields in protein structure calculation. These models involve a temperature or weighting factor that needs to be inferred from the data. Bayesian inference stipulates to determine the temperature based on the model evidence. This is challenging because the model evidence, a ratio of two high-dimensional normalization integrals, cannot be calculated analytically. We outline a replica-exchange Monte Carlo scheme that allows us to estimate the model evidence by use of multiple histogram reweighting. The method is illustrated for an Ising model and examples in protein structure determination.

[1]  I D Campbell,et al.  Some NMR Experiments and a Structure Determination Employing a {15N,2H} Enriched Protein , 1998, Journal of biomolecular NMR.

[2]  M. Nilges,et al.  Influence of non-bonded parameters on the quality of NMR structures: A new force field for NMR structure calculation , 1999, Journal of biomolecular NMR.

[3]  Jonathan Michael Pryce.  Statistical mechanics of image restoration , 1995 .

[4]  Michael Nilges,et al.  Replica-exchange Monte Carlo scheme for bayesian data analysis. , 2005, Physical review letters.

[5]  Ad Bax,et al.  Validation of Protein Structure from Anisotropic Carbonyl Chemical Shifts in a Dilute Liquid Crystalline Phase , 1998 .

[6]  Zhenyu Zhou,et al.  Approximate maximum likelihood hyperparameter estimation for Gibbs priors , 1997, IEEE Trans. Image Process..

[7]  Jun-ichi Inoue,et al.  Dynamics of the maximum marginal likelihood hyperparameter estimation in image restoration: gradient descent versus expectation and maximization algorithm. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[8]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Alan M. Ferrenberg,et al.  Optimized Monte Carlo data analysis. , 1989, Physical Review Letters.

[10]  Amos Storkey,et al.  Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics (AISTATS-12) , 2012 .

[11]  Xiao-Li Meng,et al.  Simulating Normalizing Constants: From Importance Sampling to Bridge Sampling to Path Sampling , 1998 .

[12]  Michael Nilges,et al.  Materials and Methods Som Text Figs. S1 to S6 References Movies S1 to S5 Inferential Structure Determination , 2022 .

[13]  Beale.  Exact distribution of energies in the two-dimensional ising model. , 1996, Physical review letters.

[14]  H. Kiwata,et al.  Physical consideration of an image in image restoration using Bayes’ formula , 2012 .

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

[16]  D. Baker,et al.  Design of a Novel Globular Protein Fold with Atomic-Level Accuracy , 2003, Science.

[17]  Chin-Tu Chen,et al.  Image Restoration Using Gibbs Priors: Boundary Modeling, Treatment of Blurring, and Selection of Hyperparameter , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[19]  M. Nilges,et al.  Computational challenges for macromolecular structure determination by X-ray crystallography and solution NMRspectroscopy , 1993, Quarterly Reviews of Biophysics.

[20]  Michael Nilges,et al.  Bayesian inference applied to macromolecular structure determination. , 2005, Physical review. E, Statistical, nonlinear, and soft matter physics.

[21]  Wang,et al.  Replica Monte Carlo simulation of spin glasses. , 1986, Physical review letters.

[22]  L. Goddard.  Information Theory , 1962, Nature.