The goal of the proposed research is to develop a method for the refinement of the solution conformations of proteins, as determined by the combination of nuclear magnetic resonance (NMR) and distance geometry, to higher precision than has heretofore been possible. The proposed refinement method consists of two distinct phases. In the first phase, the structures obtained from the nuclear Overhauser enhancement (NOE) data alone will be further modified so as to be consistent with additional NMR data whose geometric interpretation is ambiguous. Examples of such data include coupling constants, NOE's between nonstereospecifically assigned resonances, and ring current shifts. In the second phase of the refinement, the structures obtained in the first phase will be subjected to a sophisticated constrained energy minimization procedure, by means of which their local geometry and van der Waals packing can be improve without introducing violations of the experimental data. This refinement procedure will be tested by using the experimental NMR data which is available on proteins of known crystal structure, in an attempt to find significant differences between their solution and crystal structures. In order to establish that the observed differences lie outside the range of thermal fluctations, and to account for the effects of motion on the NMR data, we will also study the flexibility of the computed conformations by means of established techniques such as harmonic and molecular dynamics.
Showing the most recent 10 out of 13 publications
