The information content of macromolecular spectra

The object of NMR spectroscopy on proteins, nucleic acids and other complex polymers is to obtain information on their structure and dynamics. Assignment of individual spectral lines discussed in the preceding section is a prerequisite to one s ability to decipher this information. Actually to do so it is also necessary to understand the nature of the structural and dynamic information inherent in each feature of the spectrum. And one is rightfully asked is the nature of this information such that the result will be worth the labor required to obtain a significant set of assignments An important part of the answer to this question is that NMR is the ONLY physical method which can provide any information at essentially atomic resolution on the structure and dynamics of macromolecules in solution, as well as in the solid state. Whatever the limits of this information, it is better than none. Knowledge of these limits, as well as of the nature of spectroscopic information, is nevertheless necessary, both to realize expectations and to avoid conclusions that go beyond the capabilities of the method. The information content and its limits can be meaningfully discussed for each measured parameter separately, and the common features summarized at the end. 

A more quantitative interpretation of chemical shift data cannot be made at present, with few exceptions, such as the calculations of ring current shifts. The theory of chemical shifts is by and large inadequate to deal with the subtleties of the observed changes. A priori calculations remain grossly inaccurate, because the large number of parameters contained in the equations cannot be derived from any experimental measurement. Even calculations of ring current shifts involve iterative fitting of these parameters to known crystal structure data. The view of molecular structure derived from a study of chemical shifts thus reveals a wealth of detail blurred in its essentials. 

---