Re: [AMBER] 1-4 scale factors

From: Jason Swails <jason.swails.gmail.com>
Date: Wed, 23 May 2012 19:49:36 +1000

I'm jumping in a little bit late here (but better than never?).

SCEE and SCNB were removed from the input file in Amber 11. This was done
to allow "proper" 1-4 scaling for a GLYCAM/AMBER force field mix (since, as
mentioned, GLYCAM uses no scaling on 1-4 interactions and Amber uses 1.2
and 2.0 for EEL and VDW, respectively).

I would strongly suggest against changing these scaling parameters unless
you have a very good reason not to. In that case, use ParmEd from
AmberTools 12. You can set scnb and scee via the commands:

scnb 2.0
scee 1.2

You can also use xparmed.py to make these changes. This behavior is
identical to setting scnb and scee in the mdin file of Amber10 and
earlier. (Note, don't use the addDihedral command -- that adds a brand new
dihedral term).

HTH,
Jason

On Mon, May 21, 2012 at 5:31 PM, Karl N. Kirschner <
kkirsch.scai.fraunhofer.de> wrote:

> Hi Urszula (and Francois),
>
> Here is my $0.02 worth... The amount of influence that using different
> electrostatic 1-4 scaling factors has on the potential energy is dependent
> upon the magnitude of the partial atomic charges on atoms #1 and #4, which
> is connected to the charge derivation approach as mentioned below. For
> carbohydrates and analogues thereof, these partial atomic charges are
> rather negative for oxygens atom types, regardless if using the Amber or
> Glycam06 approach, and thus the dihedral potential for O-X-X-O sequence is
> sensitive to different scaling factors. The issue of 1-4 scaling factors
> comes up in carbohydrates because there is a 1-4 and a 1-5 oxygen-oxygen
> interaction in the omega angle rotation (i.e. O6---O5 and O6----O4
> interactions). During Glycam06's development, the use of 1.0 for scee and
> scnb seemed reasonable and it helped to make the dihedral parameter
> optimization and transferability better behaved.
>
> When testing the influence of different 1-4 scaling factors and charge
> sets, it is important to include an observable that is sensitive to this -
> and, in my opinion, the carbohydrate's omega angle (or analogue thereof) is
> such an observable.
>
> Best regards,
> Karl
>
> ----- Original Message -----
> From: "FyD" <fyd.q4md-forcefieldtools.org>
> To: "AMBER Mailing List" ambermd.org>
> Sent: Saturday, May 19, 2012 8:30:29 AM
> Subject: Re: [AMBER] 1-4 scale factors
>
> Urszula,
>
> Concerning applying 1-4 scale factors, the GLYCAM FF uses different
> values than AMBER. The GLYCAM FF uses also its own charge derivation
> approach compared to AMBER.
>
> We have used AMBER 1-4 scale factors following a highly consistent
> charge derivation (Amber protocole) for a cyclodextrin based
> glycopeptide and the results were more satisfactory than these
> obtained with GLYCAM.
>
> My feeling is that 1-4 scale factors depend on the charge derivation
> procedure and dihedral FF parameters (which should be re-developed for
> 'some' sugar units).
>
> Thus, you can use the 1-4 scale factors you wish if you rigorously
> validate your approach. Thus, a good learning procedure for you would
> be to run two jobs with the two types of scaling factors and study
> what's going on...
>
> regards, Francois
>
>
> > Regarding applying 1-4 scale factors, In the manual for AMBER 12 I
> > found that (pasted below)
> >
> > Usage: addDihedral
> > <scee> <scnb> []
> > The dihedrals will be set with force constant phi_k, periodicity
> > per, phase angle phase, 1-4 electrostatic
> > scaling factor scee (this must be specified - the default Amber
> > value is 1.2 and the default
> > GLYCAM value is 1.0), the 1-4 van der Waals scaling factor scnb
> > (this must be specified - the
> > default Amber value is 2.0 and the default GLYCAM value is 1.0). The
> > type is either "normal",
> > "multiterm", or "improper". "multiterm" simply means that 1-4
> > factors for that dihedral are
> > not calculated, which should be true for all but the last term in a
> > multiterm dihedral (so 1-4
> > interactions are only counted once) and in some ring systems with at
> > most 6 atoms, since that
> > could also lead to double-counting specific 1-4 interactions.
> >
> > It is still not clear for me how to apply those factors, in the
> > xleap/tleap? or its better to change it in the glycam_06g.dat file...
> > Could anyone suggest me something?
>
>
> > Urszula,
> >
> > check the LEaP manual, it should be explained there...
> >
> > regards, Francois
> >
> >
> >> In the Glycam_06g.dat file I found quite long list of SCEE and SCNB
> >> ... see below, it means that I should change the SCNB value for all
> >> of them?
> >> or should I add extra line somewhere?
> >>
> >>
> >> O2-S -N -H 1 -0.10 0.0 6. SCEE=1.0
> >> SCNB=1.0 N-Sulfates
> >> O2-S -N -CG 1 0.11 0.0 -3. SCEE=1.0
> >> SCNB=1.0 N-Sulfates
> >> 1 0.0 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 0.0 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> S -N -CG-H1 1 0.02 0.0 3. SCEE=1.0
> >> SCNB=1.0 N-Sulfates
> >> S -N -CG-CG 1 0.0 0.0 -3. SCEE=1.0
> >> SCNB=1.0 N-Sulfates
> >> 1 -1.25 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 -4.00 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> CG-OS-CG-SM 1 0.67 0.0 -3. SCEE=1.0
> >> SCNB=1.0 Thiol-linkages
> >> 1 1.90 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 1.14 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> OS-CG-SM-CG 1 1.03 0.0 -3. SCEE=1.0
> >> SCNB=1.0 (methoxymethyl)(methyl)sulfane
> >> 1 1.44 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 0.27 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> H2-CG-SM-CG 1 0.0 0.0 1. SCEE=1.0
> >> SCNB=1.0 (methoxymethyl)(methyl)sulfane
> >> SM-CG-CG-H2 1 0.28 0.0 -3. SCEE=1.0
> >> SCNB=1.0 (2,2-dimethoxyethyl)(methyl)sulfane
> >> 1 0.04 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 -0.21 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> SM-CG-CG-OS 1 0.03 0.0 -3. SCEE=1.0
> >> SCNB=1.0 (2-methoxyethyl)(methyl)sulfane
> >> 1 0.18 0.0 -2.
> >> SCEE=1.0 SCNB=1.0
> >> 1 -1.03 0.0 1.
> >> SCEE=1.0 SCNB=1.0
> >> CG-SM-CG-H1 1 0.33 0.0 3. SCEE=1.0
> >> SCNB=1.0 Dimethylsulfane
> >> CG-CG-SM-CG 1 0.32 0.0 -3. SCEE=1.0
> >> SCNB=1.0 Ethyl(methyl)sulfane
> >>
> >> /Urszula
> >>
> >> ________________________________________
> >> Von: Lachele Foley (Lists) [lf.list.gmail.com]
> >> Gesendet: Mittwoch, 16. Mai 2012 19:19
> >> An: AMBER Mailing List
> >> Betreff: Re: [AMBER] 1-4 scale factors
> >>
> >> You put them in the parameter file in the torsion section. See the
> >> torsions section in the file $AMBERHOME/dat/leap/parm/GLYCAM_06h.dat
> >> for examples. You just have to add the keyword/value pairs
> >> "SCEE=value" and "SCNB=value" (no spaces between) in the comments
> >> section of the appropriate lines.
> >>
> >>
> >> On Wed, May 16, 2012 at 7:01 AM, Urszula Uciechowska
> >> <urszula.uciechowska.chem.umu.se> wrote:
> >>> Dear Amber users,
> >>>
> >>> I would like to somehow use the 1-4 scale factors, could someone
> >>> tell me how to define it? should I do it in input file? or during
> >>> the topology file preparations?
> >>>
> >>>
> >>> | Note: 1-4 EEL scale factors were NOT found in the topology file.
> >>> | Using default value of 1.2.
> >>>
> >>> | Note: 1-4 VDW scale factors were NOT found in the topology file.
> >>> | Using default value of 2.0.
> >>> | Duplicated 0 dihedrals
> >>> | Duplicated 0 dihedrals
> >>>
> >>> /Urszula
>
>
>
> _______________________________________________
> 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
>



-- 
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
Received on Wed May 23 2012 - 03:00:02 PDT
Custom Search