Hi Mary,
There are some known geometry artifacts in the AMBER force field for
nucleic acids. If you check some of the purine base atom angles, you'll
see they aren't exactly correct either (in comparison to reference*).
These deviations are relatively small and likely due to a compromise made
when assigning atom types from the limited set that were available. This
will hopefully be addressed in the future, but for now it is suggested to
proceed with the understanding that they exist, but do not seem to be the
deciding factor in the success or failure of the force field accurately
model nucleic acids.
Best,
--Niel
* Reference data:
L. Clowney et al., Geometric Parameters in Nucleic Acids: Nitrogenous Bases,
J.Am.Chem.Soc. 1996, 118, 509-518
A. Gelbin et al., Geometric Parameters in Nucleic Acids: Sugar and
Phosphate Constituents, J.Am.Chem.Soc. 1996, 118, 519-529
Also available here:
http://ndbserver.rutgers.edu/ndbmodule/standards/ideal_geometries.html
On Sat, Nov 8, 2014 at 9:15 AM, Mary Varughese <maryvj1985.gmail.com> wrote:
> Sir,
>
> I got some valuable information. i am really thankful for the reply.
> sir, if the parm10.dat provides C1'-N9 equillibrium distance as 1.475
> angstrom why 1.52 angstrom in the pdb file i saved from xleap with ff10
> sourced (residue A: RNA Adenine)?
>
> On Fri, Nov 7, 2014 at 6:25 PM, Jason Swails <jason.swails.gmail.com>
> wrote:
>
> > On Fri, 2014-11-07 at 15:17 +0530, Mary Varughese wrote:
> > > Sir,
> > >
> > > Usually C-N bond distance is ~1.47 angstroms.
> > > In DNA and RNA, the A nucleotide has C-N (C1'-N9) distance given as
> > > 1.52 angstrom (from leap). if in a struture this distance is 1.45
> > > angstrom. does amber force it to 1.52 and would it cause any
> > > instabilty to the system? is this distance difference significant?
> >
> > Let's have a look. The C1' and N9 atom types in the 'A' residue of
> > nucleic12.lib have atom types CT and N*. If we look at parm10.dat where
> > those parameters are defined, we see the line:
> >
> > CT-N* 337.0 1.475 JCC,7,(1986),230; ADE,CYT,GUA,THY,URA
> >
> > showing that the equilibrium bond distance is 1.475 Ang and the bond
> > force constant is 337.0 kcal/mol/A^2 (this is actually half of the force
> > constant defined by Hooke's law, so the "canonical" force constant is
> > actually 674.0 kcal/mol/A^2). So the bond potential applies a force
> > between those two atoms pushing or pulling their distance toward the
> > value of 1.475 A.
> >
> > Now let's look at the Energy (E=1/2.k.(x-x0)^2) and the force
> > (F=-k(x-x0)). At a distance of 1.47 A as you asked, the bond energy
> > here becomes 0.0084 kcal/mol with a force of 3.37 kcal/mol/A. If we
> > instead look at the distance 1.52, we see that our bond energy is 0.682
> > kcal/mol and our force is 30.33 kcal/mol/A.
> >
> > This energy is reasonably small -- on the order of kT. I look at
> > energies because forces are harder to gauge whether they are "big" or
> > "small", since we don't commonly think in terms of forces when doing
> > biochemistry. You can run some other back-of-the-envelope calculations
> > (like the electrostatic force between two monovalent ions separated by
> > 10 Angstroms) to get a general idea of how "big" 30.33 kcal/mol/A is
> > compared to other forces arising in the force field.
> >
> > HTH,
> > Jason
> >
> > --
> > Jason M. Swails
> > BioMaPS,
> > Rutgers University
> > Postdoctoral Researcher
> >
> >
> > _______________________________________________
> > 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
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Nov 08 2014 - 16:00:02 PST
