Questions and problems?

Analysis questions

I just ran ANAL on my nucleic acid structure, and the h-bonding energies are zero. I know that h-bonding is present because the distances are reasonable; also, I had previously run ANAL using 4.0 - what is happening?

The new force field does not use hydrogen bonding terms. You could use distances (e.g. from Carnal) or overall base-base interaction energies to characterize your structures.

I used carnal for averaging of coordinates and obtained average interatomic diatances on a viewer. I also performed interatomic distance analysis using DIST in DECLARE. The values from the two methods were not the same. They were different in the order of 0.1 Angstromes. Why are they so different?

Consider for example a methyl group which is rotating during dynamics. If you average the coordinates, you will notice that the C-H bond distances shrink, although all in each frame the distance is ~1A. Looking at a planar analog:


                X      (Z)      X'
                 \             /
                  \           /
                   \         /
                    \       /
                     \     /
                      \   /
                       \ /
                        A

If atom X is observed at positions X and X', the average position is Z, and we can see that

      ( d(A-X) + d(A-Y) ) / 2    != d(A-Z)

The measurement of an average structure from a trajectory is not equivalent to the average of the same measurement on each set in the trajectory.

Web Masters <webadmin@www.amber.ucsf.edu>

Last modified: Mon Jan 22 19:16:14 PST 1996