if I understand, you're rotating a dihedral inside the ring, which leads to
breaking of one of the bonds. GAFF is not intended to do this (and MM in
general is not normally intended to model bond breaking or formation).
You're calculating the dihedral energy, but inside a ring what you're
seeing is the energy associated with the bond distortion needed to rotate
around a bond in a ring.
Do you expect this bond to break in your simulations? If so, you should use
a different approach. If you're just wondering about parameters for small
changes to the ring (withing normal thermal energy), then using GAFF is
probably more reasonable.
On Sun, Apr 28, 2019 at 4:10 PM Erdem Yeler <erdemyeler.gmail.com> wrote:
> Hi amber lovers,
> I am trying to generate new force field parameters based on quantum
> mechanical calculation. I used amber manual "parameter development" title
> for this and I calculated bond-angle-non bond paramters but I am having a
> hard time obtaning dihedral parameters for aromatics Let's consider about
> pyrazin. These are gaff2 pyrazin parameters :
> ca-ca-nb-ca 2 9.600 180.000 2.000
> h4-ca-nb-ca 2 9.600 180.000 2.000
> nb-ca-ca-nb 4 14.500 180.000 2.000
> h4-ca-ca-nb 4 14.500 180.000 2.000
> h4-ca-ca-h4 4 14.500 180.000 2.000
>
> When I scanned "ca-ca-nb-ca" dihedral angels, I found that dihedral bond
> is breaking at 100 degre and 111 kcal/mol energy. According to manual, if
> this is an ordinary dihedral ca-ca-nb-ca should have (111/2=)55,5
> kcal/mol V/2 value.
> But in gaff2 it is just 9.60kcal/mol. I couldn't understand how? and why?
> Beside, idivf shouln't be 4 instead of 2?
> Thank you.
> NOTE: Here is the picture.every step is degree.
> [image: image.png]
> _______________________________________________
> 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 Thu May 02 2019 - 10:00:02 PDT
