[gmx-users] Add acetyl and amide caps to a peptide with pdb2gmx

[Stephane Abel]

Hi gromacs users

I would like to add a acetyl and amide caps (N and C term) to a peptide 
for subsequent simulations with with GMX(v4.05) and the ffG53a6 ff . For 
this purpose i have added manually the missing atoms in my pdb files 
taken from the protein databank

My pdb file has the following structure

ATOM 1 C ACE A 145 17.175 5.988 69.357 1.00 0.00 C
ATOM 2 O ACE A 145 18.274 6.337 68.399 1.00 0.00 O
ATOM 3 CA ACE A 145 17.256 4.596 68.703 1.00 0.00 C
ATOM 4 N HIS A 146 16.165 6.227 70.468 1.00 44.68 N
ATOM 5 CA HIS A 146 15.557 6.451 71.779 1.00 45.28 C
ATOM 6 C HIS A 146 14.848 5.220 72.313 1.00 42.21 C
..

ATOM 65 ND2 ASN A 152 12.707 -14.633 75.307 1.00 42.30 N
ATOM 66 N NH2 A 153 14.154 -16.823 71.228 1.00 0.00 N

For that I use the following command

pdb2gmx_mpi -f beta-I-Ac-NH2.pdb -o beta-I-Ac-NH2.gro -p 
beta-I-Ac-NH2.top -ignh

I obtain the following error:

Fatal error:
atom N not found in residue 1ACE while combining tdb and rtp

What is wrong since the N atom exists in the residu ACE in my pdb file

I have also read in the mailing list that the "-ter" argument can be use 
to select the N and C term types

If i use the command

pdb2gmx_mpi -f beta-I-Ac-NH2.pdb -o beta-I-Ac-NH2.gro -p 
beta-I-Ac-NH2.top -ignh -ter

i obtain an other message "Fatal error:
Atom H not found in residue 9 while adding improper"

Thank you in advance for your help

Stefane