QUANTITATIVE-EVALUATION OF CROSS-PEAK VOLUMES IN MULTIDIMENSIONAL SPECTRA BY NONLINEAR-LEAST-SQUARES CURVE-FITTING

A procedure for quantitative evaluation of cross-peak volumes in spect ra of any order of dimensions is described; this is based on a general ized algorithm for combining appropriate one-dimensional integrals obt ained by nonlinear-least-squares curve-fitting techniques. This proced ure is embodied in a program, NDVOL, which has three modes of operatio n: a fully automatic mode, a manual mode for interactive selection of fitting parameters, and a fast reintegration mode. The procedures used in the NDVOL program to obtain accurate volumes for overlapping cross peaks are illustrated using various simulated overlapping cross-peak patterns. The precision and accuracy of the estimates of cross-peak vo lumes obtained by application of the program to these simulated cross peaks and to a back-calculated 2D NOESY spectrum of dihydrofolate redu ctase are presented. Examples are shown of the use of the program with real 2D and 3D data. It is shown that the program is able to provide excellent estimates of volume even for seriously overlapping cross pea ks with minimal intervention by the user. (C) 1995 Academic Press, Inc .