AMBER Archive (2005)

Subject: RE: AMBER: Dihedral Energy calculation

From: yen li (chem_me2000_at_yahoo.com)
Date: Wed Mar 02 2005 - 12:03:24 CST


Dear Dr. Ross,

Sorry I was not clear in defining my problem.

I am only interested in calculating dihedral energy of
a molecule(I am not interested in other energies 1-4
interactions etc).
Since I am calculating it manually I want to validate
 my result(ie total dihedral energy of the molecule)
with that of Amber.
  
As Dr. Junmei has stated I am using
0.5 * Vn * (1 + cos(n*phi - gamma))

Now for certain Atom types there are more than one
parameter in gaff.dat.

eg.

Case I
hc-c3-c -o 1 0.80 0.0 -1. Junmei et al,
1999
hc-c3-c -o 1 0.08 180.0 3. Junmei et al,
1999

Case II
c3-c3-n-c 1 0.5 180.0 -4. phi,psi,parm94
c3-c3-n-c 1 0.15 180.0 -3. phi,psi,parm94
c3-c3-n-c 1 0.53 0.0 1. phi,psi,parm94

If I encounter such atom types dihedral, should I
i) consider(ALWAYS) all the 2 sets of parameter (Case
I) and 3 sets (Case II).

ii) since I am doing it manually and there is no
problem of over counting 1-4 non-bonded interactions.
Do the sign of the multiplicity(n) affect my
calculation. What does the -ve multiplicity signify.

Thanks and regards,

Yen

--- Ross Walker <ross_at_rosswalker.co.uk> wrote:

> Dear Yen
>
> > "On the other hand, if n is a positive number, the
> > followed term is for another torsional angle" is
> not
> > clear can you please ellaborate.
> > hc-c3-c -o 1 0.80 0.0
> -1.
> > Junmei et al, 1999
> > hc-c3-c -o 1 0.08 180.0
> 3.
> > Junmei et al, 1999
>
> If I understand what you are asking it concerns why
> one torsion angle has a
> -ve term and one does not. This is due to the way
> the torsion angles are
> used internally in the code. Amber scales 1-4
> nonbond interactions by scee
> and scnb and thus requires a list of 1-4
> interactions. To calculate this
> list on every step would be very expensive but
> fortunately one already
> exists in the form of the dihedrals terms. However,
> it is possible to have
> more than one dihedral term for a given set of atoms
> and so the 1-4n
> non-bond code needs a way to avoid counting each 1-4
> non bond interaction
> more than once. This is done by marking the first
> torsion angle's n value as
> positive and the rest negative. The code for scaling
> the 1-4 interactions
> then simply only considers dihedral terms that have
> +ve n values and so
> avoids the problem of over counting. I hope this
> makes sense. In terms of
> the actual dihedral terms the -ve sign makes no
> difference to the
> calculation.
>
> > The torsions H1-C4-C5-O6, H2-C4-C5-O6, H3-C4-C5-O6
>
> > are
> > of the same type but Amber is reporting
> H2-C4-C5-O6
> > two times and the rest 3 times in the topology
> file.
> > I am trying to calculate the same. Therefore I
> need to
> > understand how this is done in amber.
>
> I'm not quite sure what you mean here, can you list
> the terms as they appear
> in the prmtop file.
>
> All the best
> Ross
>
> /\
> \/
> |\oss Walker
>
> | Department of Molecular Biology TPC15 |
> | The Scripps Research Institute |
> | Tel:- +1 858 784 8889 | EMail:-
> ross_at_rosswalker.co.uk |
> | http://www.rosswalker.co.uk/ | PGP Key available
> on request |
>
> Note: Electronic Mail is not secure, has no
> guarantee of delivery, may not
> be read every day, and should not be used for urgent
> or sensitive issues.
>
>
-----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to
> majordomo_at_scripps.edu
>

        
                
__________________________________
Celebrate Yahoo!'s 10th Birthday!
Yahoo! Netrospective: 100 Moments of the Web
http://birthday.yahoo.com/netrospective/
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu