Determination of Macromolecular Structure and Dynamics by NMR

Abstract

The possibility of defining solution structures of proteins and nucleic acids from NMR data has been advanced for many years [1, 2, 3]. In several cases, partial successes have been achieved [4, 5, 6]. However, a closer examination of the lines of argument used in the interpretation of NMR data forces one to conclude that they have not led to the development of an independent, generally applicable method for this purpose. The most successful examples (5, 6) have relied heavily on prior knowledge of the crystal structure — or of very similar crystal structures — and/or on a priori additional constraints, such as the assumption that a single rigid structure exists in solution which can be calculated from the data by minimizing either an error function or the energy of the structure. The danger of such procedures has been discussed elsewhere [8, 9, 10].