[gmx-users] Conversion of improper angles from amber object file to gromacs

[Daniel Bauer]

Hi,

I  am trying to convert forcefield parameters of zwitterionic glycine
from an amber object file format to gromacs .rtp (to use it with
amber99sb, which lacks zwitterionic amino acids).

The amber library file I plan to use can be found here:
https://personalpages.manchester.ac.uk/staff/Richard.Bryce/amber/pro/zaa.off

This is what i have so far:

[ZG]
 [ atoms ]
     N    N3          -0.408600     1
    H1    H            0.300300     2
    H2    H            0.300300     3
    H3    H            0.300300     4
    CA    CT          -0.026400     5
   HA1    H1           0.073800     6
   HA2    H1           0.073800     7
     C    C            0.835500     8
     O    O2          -0.724500     9
   OXT    O2          -0.724500    10
 [ bonds ]
     N    H1
     N    H2
     N    H3
     N    CA
    CA   HA1
    CA   HA2
    CA     C
     C     O
     C    OXT

However, I am unsure about the correct conversion of impropers. In the
.off file format, those are based on the atom type, while gromacs
expects them to be based on atom names. This leaves room for some
ambiguity when the same atom type is used multiple types (i.e the COO or
NH3 group).

The improper parameter section of the .off file looks like this:

!entry.zaa_improper.parm.torsions table  str type1  str type2  str
type3  str type4  int type  dbl kp  int n  dbl p0  str desc
 "CT" "O2" "C" "O2" 1 10.500000 2 3.141594 ""
 "CT" "N" "C" "O" 1 10.500000 2 3.141594 ""

Now, for the CT-O2-C-O2 improper, this would map to two possible
combinations:

  [ impropers ]
  ;  atom1 atom2 atom3 atom4 phase(deg) kd(kJ/mol-1rad-2) dn
     CA    O    C     OXT    180        43.932            2
     CA    OXT    C     O    180        43.932            2

Are both entries required in the gromacs [ improper ] section or must I
put only one of them?


Thanks,

Daniel

-- 
Daniel Bauer, M.Sc.

TU Darmstadt
Computational Biology & Simulation
Schnittspahnstr. 2
64287 Darmstadt
bauer at cbs.tu-darmstadt.de