[gmx-users] GB parameter(s) missing or negative but they aint missing nor negative

[Gert Peters]

Dear all,

Im trying to run an md or em using an implicit solvation method using 
gromacs 4.5.5 but I allways get the same errors.

GB parameter(s) missing or negative for atom type 'OS'

GB parameter(s) missing or negative for atom type 'H2'

GB parameter(s) missing or negative for atom type 'N*'

GB parameter(s) missing or negative for atom type 'CK'

GB parameter(s) missing or negative for atom type 'P'

-------------------------------------------------------
Program grompp, VERSION 4.5.5
Source code file: grompp.c, line: 1123

Fatal error:
Can't do GB electrostatics; the implicit_genborn_params section of the 
forcefield is missing parameters for 5 atomtypes or they might be negative.

Istart from default.pdb (41bp RNA) and perform following:

pdb2gmx -f default.pdb -ff amber03 -water none -ignh
editconf -f conf.gro -d 2 -bt triclinic -c
grompp -f em_gbis.mdp -c out.gro

with em_gbis.mdp containing:

integrator= steep
nsteps= -1
nstlist= 10
emstep=1
rlist= 2.0
coulombtype= cut-off
rcoulomb= 2.0
vdw-type= cut-off
rvdw= 2.0
nstenergy= 10
implicit_solvent = GBSA
constraints=none
rgbradii=2.0
gb_algorithm=OBC
sa_algorithm=Ace-approximation

Now the odd part is that I added all missing parameters to the 
/gromacs-4.5.5/share/top/amber...ff/gbsa.itp for example for H2:

...
H4           0.1      1      1        0.115 0.85 ; H
H5           0.1      1      1        0.125 0.85 ; H
H2           0.1      1      1        0.125 0.85 ; H
...

I tried doing this for amber 94, 99 and 03 but allways the same result.

Anyone got a clue? I dont think I did something wrong as the implicit 
solvation model previously worked. All input is appreciated.

Thanks in advance,
Gert