The Rowland Institute at Harvard

Biophysical Chem Lab

Members

Jeff Hoch
Alan Stern

Research

LSIII
Human Prolactin

NMR Data Processing
Macromolecular
  Structure Calculations

Group Alumni

Peter Connolly
Claudio Chuaqui
Eric LaRosa
Kuo-Bin Li
George Maalouf
John Osterhout

Rowland Home

NMR Data Processing

Our laboratory has broad interests in the area of NMR data processing, which are manifested in a software package, the Rowland NMR Toolkit, and a recently published book NMR Data Processing. See the Toolkit Web page for information on obtaining the software.

An area of particular interest is modern, or non-Fourier methods for spectrum analysis. While fast Fourier algorithms played a vital role in the development of modern NMR spectroscopy, the discrete Fourier transformation suffers from several limitations, among them difficulty obtaining high resolution spectral estimates from short data records, and the requirement that data be collected at uniformly spaced intervals. For decaying time domain signals, typical of NMR data, this latter constraint means that valuable data acquisition time must be devoted to portions of the signal where the signal-to-noise ratio is low. Modern spectrum analysis methods, such as maximum entropy reconstruction, are not similarly constrained, and data can be sampled at arbitrary intervals. Thus sampling can be weighted toward portions of the signal where S/N the highest, decreasing data acquisition times, without loss of resolution, or improving S/N for the same data acquisition time.

MaxEnt is more difficult to compute the DFT, and by its very nature is nonlinear. We have developed efficient algorithms for computing MaxEnt reconstructions, including parallel codes for symmetric multiprocessing and distributed-parallel computer architectures. We have also developed methods for compensating for the nonlinearity of MaxEnt reconstruction, and we continue to explore strategies for exploiting nonlinear sampling in order to reduce data acquisition times and to improve spectral quality, in close collaboration with Gerhard Wagner's group at Harvard Medical School.

Relevant publications

"NMR Data Processing"
Jeffrey C. Hoch and Alan S. Stern
Wiley-Liss, New York (1996)

"Quantification of Maximum-Entropy Spectrum Reconstructions"
Peter Schmieder, Alan S. Stern, Gerhard Wagner, and Jeffrey C. Hoch
J. Magn. Reson.,125, 332-339 (1997)

"Improved resolution in triple-resonance spectra by non-linear sampling in the constant time domain"
Peter Schmieder, Alan S. Stern, Gerhard Wagner, and Jeffrey C. Hoch
J. Biomol. NMR, 4, 483-490 (1994)

"Application of non-linear sampling schemes to COSY-type spectra"
Peter Schmieder, Alan S. Stern, Gerhard Wagner, and Jeffrey C. Hoch
J. Biomol. NMR, 3, 569-576 (1993) "A New, Storage-efficient Algorithm for Maximum Entropy Spectrum Reconstruction"
Alan S. Stern and Jeffrey C. Hoch
J. Magn. Reson., 97, 255-270 (1992).

"Maximum Entropy and The Nearly Black Object"
David L. Donoho, Iain M. Johnstone, Jeffrey C. Hoch, and Alan S. Stern
J. R. Statist. Soc. B54, 41-81 (1992).

"Does the maximum entropy method improve sensitivity?"
David. L. Donoho, Iain M. Johnstone, Alan S. Stern, and Jeffrey C. Hoch
Proc. Natl. Acad. Sci. USA, 87, 5066 (1990).

"Maximum Entropy Reconstruction of Complex (Phase-Sensitive) Spectra"
Jeffrey C. Hoch, Alan S. Stern, David L. Donoho, and Iain M. Johnstone
J. Magn. Reson., 86, 236 (1990).

"Maximum Entropy Reconstruction in NMR"
Jeffrey C. Hoch and Alan S. Stern
in "Encyclopedia of NMR", D.M. Grant and R.K. Harris, eds., John Wiley & Sons, Chichester (1996).

"Modern Spectrum Analysis in NMR: Alternatives to FT"
Jeffrey C. Hoch, Meth. Enzym., 176, 216-241 (1989).

Copyright © 2002 The Rowland Institute at Harvard.

Last modified Thursday, March 9, 2006.