AMBER Archive (2007)

Subject: Re: AMBER: RE: dihedral term charge dependence

From: Carlos Simmerling (carlos.simmerling_at_gmail.com)
Date: Mon Apr 02 2007 - 11:14:15 CDT

the dihedral term is fit as the *difference* between MM and QM energies.
As such, it explicitly includes the charges and a different charge set
would very likely lead to different dihedral terms. In my opinion they
cannot be swapped between different charge models. However, some
charge models may be close enough that it doesn't make a big difference.
The point is that there is no reason to expect that they are interchangable.

On 4/2/07, Williamson, Mark <mark.williamson_at_imperial.ac.uk> wrote:
>
>
>
> Hi, sorry for the delay in the reply. I've cc'ed this response to the list,
> since I think it may be useful.
>
>
> >Frankly speaking, I'm confused because I've thought that dihedral term
> >is independent of charges values (at least in ff 94-99, I've not used
> >yet polarizable ff). It follows from AMBER ff equation
> http://amber.scripps.edu/eqn.txt
> >Well, may be I misunderstood something. Would you be so kind to
> >explain me what you meant.
>
>
> If you look in the AMBER9 manual:
> http://amber.ch.ic.ac.uk/doc9/amber9.pdf
>
> and search for the term SCEE. This is the 1-4 electrostatic scaling factor
> and the electrostatic interaction between 1-4 atoms is divided by this
> amount. One can gain explicit information about how the value is used in
> sander by grepping for scee in $AMBERHOME/src/sander/ene.f .Also, there is
> also a force field specific scaling factor for 1-4 VDW interactions as well;
> SCNB
>
> Hence, the dilhedral term is not an exclusive energy term for 1-4
> interactions, and I'd agree with you that the equation at
> http://amber.scripps.edu/eqn.txt is misleading with respect
> to the dihedral term. For very old force fields, the the SCEE value used to
> be set to 2.0, but now (the default) is 1.2 for FF94. Be aware that there
> may be issues in using two force fields of different SCEE's (and SCNB's)
> within the same simulation, since there is no way to "tell" sander to treat
> each parameter set with its own SCEE value.
>
> Professor Thomas E. Cheatham speaks about this a bit in the following post:
> http://amber.ch.ic.ac.uk/archive/200505/0204.html
>
>
> Regards,
>
> Mark 

--
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu