[gmx-users] Problem in atom name correspondence between protein topology and coordinates after mutation grompp

[himanshu khandelia]

Hi Folks,

I am not being able to debug a little problem. I hope someone will be
able to point out the mistake. Here is what I am trying to do:

I need to mutate a SER to a GLU residue in a protein. However, after
mutation (and making a new topology), when I run grompp, I get the
following error:

######################
processing coordinates...
Warning: atom names in protein.top and p.gro don't match (CD2 - HD1)
Warning: atom names in protein.top and p.gro don't match (CE1 - CD2)
Warning: atom names in protein.top and p.gro don't match (NE2 - CE1)
Warning: atom names in protein.top and p.gro don't match (HE2 - NE2)
WARNING 2 [file "protein.top", line 35]:
  4 non-matching atom names
  atom names from protein.top will be used
  atom names from final.gro will be ignored
######################

So, the order or names of atoms do not correspond correctly any more.
I cannot find out why this is the case. Here are the steps that I used
to make the mutation:

######################
1. Convert the initial structure (init.gro) to a pdb (init.pdb) using editconf
2. Pull out coordinates of the SER  from init.pdb and change residue
name from SER to GLU. Lets call this file residue-to-be-mutated.pdb
3. Rename the following atoms: (a) OG to CG (b) HG to CD in
residue-to-be-mutated.pdb
4. Guess remaining coordinates (for the 2 oxygen atoms on the carboxyl
group) in some external software
5. plug in the guessed coordinates into residue-to-be-mutated.pdb
6. Removed the coordinates of SER from init.pdb, and replaced them by
coordinates in residue-to-be-mutated.pdb. Note that because of the
insertion of two extra atoms, the numbering of atoms in the pdb file
is not continuous any more.
7. Convert the resulting pdb file into .gro format using editconf.
Final output is final.gro, which should have continuous numbering of
atom numbers.

8. Then I also make a new protein topology using pdb2gmx.
######################

Any suggestions, anyone ?

- The number of atoms in the initial and final .gro files differ by
two, which is the same as the difference in the number of atoms
between a SER and a GLU amino acid