[gmx-users] questions about Ryckaert-Bellemans type dihedrals

[Yunhui Ge]

Hi,

I have a question regarding Ryckaert-Bellemans dihedral potentials
(function type 3). In the manual of GROMACS 4.5.4 it says to use
Ryckaert-Bellemans type, one needs to erase the 1-4 interaction in [pairs]
section. My first question is what will happen if I use this potential
(type 3) but leave the 1-4 interaction in [pairs] section?
Next in the newer version GROMACS (5.1.4) the manual says:

*In the original implementation of the potential for alkanes no 1-4
interactions were used, which means that in order to implement that
particular force field you need to remove the 1-4 interactions from the [
pairs ] section of your topology. In most modern force fields, like OPLS/AA
or Amber the rules are different, and the Ryckaert-Bellemans potential is
used as a cosine series in combination with 1-4 interactions.*

What force field is considered as most modern force fields? (I'm using
AMBER ff99SB-ildn-nmr.) Again what will happen if I don't remove those 1-4
interactions in the [pairs] section?
My last question is should I expect huge differences between
Ryckaert-Bllemans potential (type 3) and periodic (type 9) in simulations
of the same molecule? I plan to run some simulation to test that but I
wonder if there is a general answer based on the previous experience. Based
on the manual I know RB potential excludes 1-4 interactions whereas
periodic type include 1-4 interactions. I'm not clear about the performance
difference between these two dihedral types. Also if RB potential requires
the exclusion of 1-4 interactions why the manual says in modern force
fields it is used in combination with 1-4 interactions.
I'm familiar about this part and I do hope to learn more about this. Thank
you for your time in advance!

Best,

Yunhui

   1.