Atomtype not found, but atom is in FF

GROMACS version: 2021.3
GROMACS modification: Yes
Hi,

I am trying to do a relative binding energy w/ ligand modification. The ligand is a short peptide, and I have put the non-canonical amino acid into the FF. The non-canonical amino acid is trifluoromethyl-phenylalanine.

I am using a modified amber99sb FF. I have modified aminoacids.rtp (and.hdb), atomtypes, nonbonded, and bonded, all to include F3 of the trifluoromethyl group. I will put some of the lines below for context:

aminoacids.rtp:

[ TFP ] ; 4-(Trifluoromethyl)-phenylalanine
[ atoms ]
N N -0.4700 1
H H 0.3100 2
CA CT 0.0700 3
HA H1 0.0900 4
CB CT -0.1800 5
HB1 HC 0.0900 6
HB2 HC 0.0900 7
CG CA 0.0000 8
CD1 CA -0.1150 9
HD1 HA 0.1150 10
CD2 CA -0.1150 11
HD2 HA 0.1150 12
CE1 CA -0.1150 13
HE1 HA 0.1150 14
CE2 CA -0.1150 15
HE2 HA 0.1150 16
CZ CA -0.1400 17
CH CF3 1.1600 18
FH1 F3 -0.3400 19
FH2 F3 -0.3400 20
FH3 F3 -0.3400 21

atomtypes.atp:

F3 18.99800 ; fluorine, trifluoro

ffbonded.itp:

[ bondtypes ]
CT F3 1 0.13600 361990.5
[ angletypes ]
F3 CT F3 1 106.0810 986.42
CA CT F3 1 112.2780 693.14
[ dihedraltypes ]
| F3| CT| CA|CA| 9| 0.00| 0.4184| 3

ffnonbonded.itp:

[ atomtypes ]
F3 9 18.99800 0.000 A 0.2850875898 0.405848
[ pairtypes ]
CT F3 1 0.311814551349 0.130309939452

lig.top (this is how I have described the atom in the topology):

176 F3 150 TFP FH1 176 -0.34 18.998 MCH3 0.000000 18.9888
177 F3 150 TFP FH2 177 -0.34 18.998 MCH3 0.000000 18.9888
178 F3 150 TFP FH3 178 -0.34 18.998 MCH3 0.000000 18.9888

I can solvate the molecule, but as soon as I try the following prompt, I get the following response:

Command line:
gmx grompp -f tests/ions.mdp -po tests/ions.out.mdp -c tests/lig_solvated.gro -p tests/lig.top -o tests/lig_ions.tpr -maxwarn 1

Setting the LD random seed to 2122252150

Generated 2145 of the 2145 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5

Generated 2145 of the 2145 1-4 parameter combinations

ERROR 1 [file lig.top, line 214]:
Atomtype F3 not found


Program: gmx grompp, version 2021.3-MODIFIED
Source file: src/gromacs/gmxpreprocess/toppush.cpp (line 1409)

I don’t understand why this atom type can’t be found, I thought I had described it in all the relevant fields. Thank you in advance for your help.

Hi,
can you check that the correct force field directory (the modified one) is included at the begin of the topol.top? Note that if the modified force field directory is not in the working directory or in the default location, you have to include whole path.
\Alessandra

1 Like

Hi Alessandra,

Thank you for your reply. I think that may have been the problem. I remade the forcefield with a different atom name and made sure about the FF directory being mentioned in the topology. I appreciate your help.

Edan