Hi all,
I am currently finishing up a structure in X-PLOR with alternating cycles
of positional and B-factor refinement (also including a solvent mask).
I would like to then 'move' the structure to TNT for a final set of
least-squares cycles, and would like general advice on doing so. I do have
a number of specific questions, though:
1) One piece of advice I've heard is to reset all of the B-factors to the
average B and let TNT work from there. Should I? It seems like much of
the work that X-PLOR did to reach the local minimum was expended in getting
good B-factors, and I'd be knocking it out of a local minimum by resetting
them all to 23.5 A^2.
2) Should I do cycles of refinement of just XYZ of the structure before
allowing the B-factors to refine?
3) I find that an awful lot of apparent waters that are fine in X-PLOR
will blow up once I go to TNT. Is it that I'm not letting the B-factors
converge in X-PLOR and the waters aren't really there, or is it something else?
(As for the maps, sometimes waters of B=100 still show up, although that's
kinda rare.)
Thanks in advance for your input,
Yoram
--
_____________________________________________________________________________
Yoram A. Puius Albert Einstein College of Medicine
puius at aecom.yu.edu Department of Biochemistry
M.D.-Ph.D. Class of 1999 [not!] Bronx, NY 10461
_____________________________________________________________________________
Well, do you ever get the feeling / that the story's too damn real /
and in the present tense? - Ian Anderson
