Sorry about the comment, folks. We forgot to remove it. Just go
ahead and use the parameters. I have removed the comment from the
online version. I didn't change the file version to "06h-2" because
that seems silly. Please say if you disagree.
On Wed, Jan 16, 2013 at 2:03 PM, ABEL Stephane 175950
<Stephane.ABEL.cea.fr> wrote:
> I agree with your explanation Jason. My question came from the "strange" comment added after the Ho parameters term in the GLYCAM06-h1 dat file
>
> Best
>
> Stephane
>
>
> De : Jason Swails [jason.swails.gmail.com]
> Date d'envoi : mercredi 16 janvier 2013 18:06
> À : AMBER Mailing List
> Objet : Re: [AMBER] LJ parameters for Ho in GLYCAM06 changes
>
> On Wed, Jan 16, 2013 at 8:05 AM, ABEL Stephane 175950
> <Stephane.ABEL.cea.fr>wrote:
>
>> Dear all,
>>
>> I have noticed that the LJ values for Ho atom in GLYCAM06 have been
>> changed between the GLYCAM06-h-1.dat and GLYCAM06-g-1.dat force field
>> versions
>>
>> GLYCAM06-h1: Ho 0.2000 0.0300 M.B. Tessier 2011,
>> use by permission only
>> GLYCAM06-g1: Ho 0.0000 0.0000 OPLS Jorgensen,
>> JACS,110,(1988),1657
>>
>> It is possible to have in few words a justification of this change and its
>> impact in ND simulation
>>
>
> Without any VDW parameters on that hydrogen, some people were observing
> occasional blowups of their simulations caused when an atom collapsed into
> one of these hydrogens. The idea was that the connected oxygen's VDW
> radius would be large enough to block any atom from collapsing into the
> hydrogen, but sugars are tricky beasts (as the Glycam developers can well
> attest).
>
> Since 1-2 and 1-3 interactions are excluded, I think it's possible for an
> atom forming a dihedral with the hydrogen (and therefore an angle with the
> oxygen) to collapse into the Ho atom without ever feeling the repulsion
> from the oxygen. Therefore, the solution devised was to give that H atom a
> small van der Waals term. It is so small that it is virtually
> non-attractive, but still has the 1/r^12 term to prevent catastrophic
> collapses that blow up simulations.
>
> I would expect no adverse affects from such a change (and this change is
> probably documented somewhere, in an article maybe).
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> Quantum Theory Project,
> University of Florida
> Ph.D. Candidate
> 352-392-4032
> _______________________________________________
> 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 Jan 16 2013 - 15:00:02 PST
