Hello Dong--

>   I'm now refining a G-quadruplex structure using a modified refine.py
> based on the script in dir egiput/dna_refi of xplor-NIH. The input data
> is only noe constraints.
>
> My question is if I use Lennard-Jones and electrostatics and add the lines
> in refine.py:

It's almost certainly a bad idea to turn on electrostatics during
structure calculations in the absence of solvent, and perhaps
counter-ions, in the case of nucleic acids. For the case of nucleic
acids, the ORIE position/orientation database potential has been shown
to capture many structural features without the need to employ a full
force field. Now, one can *evaluate* and perhaps even do local
gradient minimization using L-J and electrostatic potentials.

Please contact me if you have further need for using realistic
nonbonded energies.

> ...
>
> The output is fine.But the program will give the message:
> InternalDynamics::step: large timestep detected. Halving.

Occasionally getting these messages is fine, but if you get a whole
bunch and the stepsize stays a ridiculously low value (say 1e-8ps)
there's clearly a problem.

best regards--
Charles
--
Charles Schwieters     email:   [email protected]
                       www:     http://schwieters.org/cds
phone: (301) 402-4914  PGP key: http://schwieters.org/cds/pgp.txt

Attachment: pgp0g0YM4ihk5.pgp
Description: PGP signature

_______________________________________________
Xplor-nih mailing list
[email protected]
http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih

Reply via email to