A Metropolis Monte Carlo implementation of bayesian time-domain parameter estimation: application to coupling constant estimation from antiphase multiplets.

The Bayesian perspective on statistics asserts that it makes sense to speak of a probability of an unknown parameter having a particular value. Given a model for an observed, noise-corrupted signal, we may use Bayesian methods to estimate not only the most probable value for each parameter but also their distributions. We present an implementation of the Bayesian parameter estimation formalism developed by G. L. Bretthorst (1990, J. Magn. Reson. 88, 533) using the Metropolis Monte Carlo sampling algorithm to perform the parameter and error estimation. This allows us to make very few assumptions about the shape of the posterior distribution, and allows the easy introduction of prior knowledge about constraints among the model parameters. We present evidence that the error estimates obtained in this manner are realistic, and that the Monte Carlo approach can be used to accurately estimate coupling constants from antiphase doublets in synthetic and experimental data.

[1]  W. Hutton,et al.  Comparison of Fourier and Bayesian analysis of nmr signals. I. Well-separated resonances (the single-frequency case) , 1992 .

[2]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[3]  P E Wright,et al.  Use of chemical shifts and coupling constants in nuclear magnetic resonance structural studies on peptides and proteins. , 1994, Methods in enzymology.

[4]  C. Dobson,et al.  Analysis of phi and chi 1 torsion angles for hen lysozyme in solution from 1H NMR spin-spin coupling constants. , 1991, Biochemistry.

[5]  William H. Press,et al.  The Art of Scientific Computing Second Edition , 1998 .

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

[7]  R. Freeman,et al.  Accurate measurement of coupling constants by J doubling , 1992 .

[8]  K OrJ.  Numerical Bayesian methods applied to signal processing , 1996 .

[9]  G. Bodenhausen,et al.  Principles of nuclear magnetic resonance in one and two dimensions , 1987 .

[10]  J. Besag,et al.  Bayesian Computation and Stochastic Systems , 1995 .

[11]  Joseph B. Rosenbach,et al.  Mathematical Tables , 1957, Nature.

[12]  An evaluation of least-squares fits to COSY spectra as a means of estimating proton—proton coupling constants. I. Simulated test problems , 1994, Journal of biomolecular NMR.

[13]  J. Prestegard,et al.  Measurement of vicinal couplings from cross peaks in COSY spectra , 1989 .

[14]  Gene H. Golub,et al.  Matrix computations , 1983 .

[15]  R. Chylla,et al.  Improved frequency resolution in multidimensional constant-time experiments by multidimensional Bayesian analysis , 1993, Journal of biomolecular NMR.

[16]  G. L. Bretthorst.  Bayesian ANALYSIS. IV. Noise and computing time considerations , 1991 .

[17]  J. Cavanagh.  Protein NMR Spectroscopy: Principles and Practice , 1995 .

[18]  Paul Bratley,et al.  A guide to simulation , 1983 .

[19]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[20]  H. Jeffreys,et al.  Theory of probability , 1896 .

[21]  J. Kiefer.  Introduction to statistical inference , 1987 .

[23]  Timothy F. Havel,et al.  An evaluation of least-squares fits to COSY spectra as a means of estimating proton—proton coupling constants II. Applications to polypeptides , 1994, Journal of biomolecular NMR.

[24]  K. Vines,et al.  Improved resolution of spectral splittings via bayesian spectral analysis , 1992 .

[25]  F. Young.  Biochemistry , 1955, The Indian Medical Gazette.

[26]  G. Larry Bretthorst,et al.  Bayesian analysis. II. Signal detection and model selection , 1990 .

[27]  Piero Barone,et al.  Bayesian estimation of parameters of a damped sinusoidal model by a Markov chain Monte Carlo method , 1997, IEEE Trans. Signal Process..

[28]  C. Robert.  The Bayesian choice : a decision-theoretic motivation , 1996 .

[29]  G. L. Bretthorst.  Bayesian analysis. V : Amplitude estimation for multiple well-separated sinusoids , 1992 .

[30]  R. Chylla,et al.  Theory and application of the maximum likelihood principle to NMR parameter estimation of multidimensional NMR data , 1995, Journal of biomolecular NMR.

[31]  G. L. Bretthorst.  Bayesian analysis. I. Parameter estimation using quadrature NMR models , 1990 .

[32]  G. Larry Bretthorst,et al.  Bayesian analysis. III. Applications to NMR signal detection, model selection, and parameter estimation , 1990 .

[33]  F. A. Seiler,et al.  Numerical Recipes in C: The Art of Scientific Computing , 1989 .

[34]  J. Prestegard,et al.  NMR STUDIES OF A 13C, 15N-LABELED GM4-LACTAM GLYCOLIPID AT AN ORIENTED MODEL-MEMBRANE INTERFACE , 1996 .