We had to add very small hydroxyl H vdW parameters in GLYCAM. Without
them, once in a while, if everything was just right, the hydroxyl H
could collapse onto a nearby atom with a negative partial charge (not
the directly bonded hydroxyl O, but some other atom nearby).
The params we added were very small, so that they don't interfere with
other properties, such as hydrogen bond formation.
You need to think carefully about the vdW terms you are adding. Are
they really necessary in your situation? Will their addition cause
other parts of the force field not to function?
To see the parameters we set, go to glycam.org/params and look at
GLYCAM_06h-1.dat.
On Wed, Apr 10, 2013 at 7:41 PM, Bill Ross <ross.cgl.ucsf.edu> wrote:
> Do you have SHAKE on for bonds with H? The H is supposed to fall
> within the vdw of the O, and without SHAKE on, the H can stretch
> beyond the O. That said, there are times when maybe an atom can
> push into the O enough to touch the H and blow things up.
>
> Bill
>
> Morteza Chehel Amirani <chehelam.ualberta.ca> wrote:
>
>> Hello
>>
>> I'm doing a MM optimization using Amber FF in Gaussian, but it fails. I
>> have HO atoms in the system and as it's defined in Amber, vdW parameters
>> for HO atom are zero. When I change the vdW parameters, the problem get
>> fixed. I'm just wondering whether or not it's admissible to do so and if it
>> is, what values for the vdW parameters I should consider?
>>
>> Sincerely
>> *Morteza Chehel Amirani*
>> PhD Candidate
>> Department of Mechanical Engineering
>> University of Alberta
>> 6-29 Mechanical Engineering Building
>> Edmonton, Alberta, T6G 2G3
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
:-) Lachele
Lachele Foley
CCRC/UGA
Athens, GA USA
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 10 2013 - 17:30:03 PDT
