From structure to chemical shift and vice-versa

Computational methods to allow quantitative exploitation of proton chemical shift data in determination of the structure of proteins was developed. A program was developed that generates a large number of conformations, and for each structures the CIS values were estimated. A genetic algorithm was used in a continuous refinement of the 1 H chemical shifts inorder to find the structure that matches the CIS values. The results show that the structures of dimer motifs characterized in solution agrees with the structures of dimers found in the x-ray crystal structures.