[gmx-users] A question about dihedral angles

[Mark Abraham]

On 11/02/10 06:52, Amir Marcovitz wrote:
> Hi all,
> i think my question is kind of trivial but i'll ask it any way:
> suppose you have atoms that are arranged on the XY plane in a square
> lattice arrangment with a spacing of 1. you bond them , add angles etc.
>   and now you want to add proper dihedral angles using function 1.
> What i know, is that the dihedral angle 'phi' between a quartet of atoms
> i,j,k,l is the angle between the two surface i,j,k and j,k,l. (i think
> this is also written in the manual..)
> yet, i experience some problems with a system similar to the one i
> described  above - i.e., the planar geometry of the atoms gets twisted
> until it blows up which make me confused and think that i may did
> some mistakes in defining these angles in the  [ dihedrals ] section of
> the topology file.

I think this is all covered in chapter 4 of the manual.

Mark

> it could realy be useful if someone can tell me the value of the
> dihedral phi for the following quartets (given the x,y coordinates):
> 1)   i (0,0)   ,  j (0,1) , k (1,1)  ,  l(1,2)
> 2)   i(0,0)   ,   j (1,0)   , k(2,0)  , l(3,0)
> 3)    i(0,0)  , j(1,0)   ,  k (2,0),    l(2,1)
> 4 )  i(0,0)   , j(1,0)  ,  k(1,1)   , l(0,1)
> i think that 1 is 180 degrees , and 2,3,4 are 0 degrees am i right?   is
> there a difference between 0 and 180?
> (i used a multipilicty of 2)
> i.e., it looks like
>
> [ dihedrals ]
>
> ; ai aj ak al funct phi cp mult
>
> 7 1 2 3        1 0.000000e+00 3.347200e+01 2.000000e+00
>
> 14 8 9 15    1 1.800000e+02 3.347200e+01 2.000000e+00
>
>   and so on..
> sorry for the bizzare question..
>