Hello,
I have been using a standard annealing script with TAD from CNS 1.0 that
included sani restraints (only 1 class, HN). Most of the structures had
no sani violations larger than 1 nor 0.1, as listed in the header of pdb
file. However, when I loaded the structures in vmd-xplor and displayed
the dipolar couplings (using the same DFS, Da and R values as for the
calculations), there were usually several violations larger than 1 or
even 2. I was quite surprised by this discrepancy. :( I listed all the
rdc, as calculated by CNS, using the sani print statement (with
threshhold = 0) and indeed, the values differed quite a bit between CNS
and vmd-xplor. I thought that this was perhaps some peculiarity of CNS,
but a sani print statement in xplor-nih (version 2.9.2) gave me
calculated values of rdcs identical to those from CNS, and different
from vmd-xplor!!!
This does not look quite right, does it? Since I was getting two
different sets of calculated rdcs, I wrote my own program to calculate
them from pdb files with ANI residue. My calculated values agree very
well with vmd-xplor values. This suggests that rdcs calculated by the
sani statement in xplor are not quite correct - which makes my very
suspicious about any sani calculations!!! If calculated rdcs are not
quite correct in xplor, then the whole sani energy term may not be doing
a good job! Or how else would it be possible to get different calculated
rdc values from the same Da, R and coordinates?
Please help,
Klara
