[gmx-users] topology and rtp file

[Justin A. Lemkul]

irene farabella wrote:
> Hello Gmx Usrs,
> 
> I am new to Gromacs. I am trying to simulate a protein-membrane
> complex using  berger-oplsaa combination . I have created a topology

Have you properly modified the Berger parameters for use with OPLS-AA? 
Otherwise, the results you will get will be nonsense :)

> file for the protein using pdb2gmx in order to then build up the
> topology for my system using the include file mechanism.
> After that I tried to use my system.top in grompp ( grompp -f xx.mdp
> -c system.pdb -p system.top -o out.tpr ) but I get a warning about
> atoms in top and pdb does not match. I went trough the mailing list

Is system.pdb the cleaned up output from pdb2gmx?

It is also better to post the actual error message.  The topology and coordinate 
file may not match in terms of atom names or in terms of the number of 
coordinates, which are both very different problems.  I guess you are referring 
to a naming mismatch?

> but I have still doubt. I checked where were the differences. It seems
> that the order of the aromatic ring in the pdb and in the top didn't
> match.
> 
> To be clear (tyr as example but the same problem occur with his):
> 
> .pdb  .top
>  N         N
>  HN        H

If this is what's in your .pdb file, then you aren't using the pdb2gmx output, 
are you?

>  CA       CA
>  HA       HA
>  CB       CB
>  HB1     HB1
>  HB2     HB2
>  CG       CG
>  CD1     CD1
>  HD1     HD1
>  CE1     CD2
>  HE1     HD2
>  CZ      CE1
>  OH      HE1
>  HH      CE2
>  CD2     HE2
>  HD2      CZ
>  CE2      OH
>  HE2      HH
>  C            C
>  O           O
> 
> Is this because the .top seems follow the order presents in the rtp
> file that I have (am I right?).

Yes, pdb2gmx will output according to the .rtp entry.

> Assuming that this is the case I am wondering if changing the atoms
> order in the rtp from:
> 
> [ TYR ]
>  [ atoms ]
>     N    opls_238   -0.500     1
>     H    opls_241    0.300     1
>    CA    opls_224B   0.140     1
>    HA    opls_140    0.060     1
>    CB    opls_149   -0.005     2
>   HB1    opls_140    0.060     2
>   HB2    opls_140    0.060     2
>    CG    opls_145   -0.115     3
>   CD1    opls_145   -0.115     4
>   HD1    opls_146    0.115     4
>   CD2    opls_145   -0.115     5
>   HD2    opls_146    0.115     5
>   CE1    opls_145   -0.115     6
>   HE1    opls_146    0.115     6
>   CE2    opls_145   -0.115     7
>   HE2    opls_146    0.115     7
>    CZ    opls_166    0.150     8
>    OH    opls_167   -0.585     8
>    HH    opls_168    0.435     8
>     C    opls_235    0.500     9
>     O    opls_236   -0.500     9
> 
> to:
> 
> [ TYR ]
>  [ atoms ]
>    N    opls_238   -0.500     1
>    H    opls_241    0.300     1
>   CA    opls_224B   0.140     1
>   HA    opls_140    0.060     1
>   CB    opls_149   -0.005     2
>  HB1    opls_140    0.060     2
>  HB2    opls_140    0.060     2
>   CG    opls_145   -0.115     3
>  CD1    opls_145   -0.115     4
>  HD1    opls_146    0.115     4
>  CE1    opls_145   -0.115     6
>  HE1    opls_146    0.115     6
>   CZ    opls_166    0.150     8
>   OH    opls_167   -0.585     8
>   HH    opls_168    0.435     8
>  CE2    opls_145   -0.115     7
>  HE2    opls_146    0.115     7
>  CD2    opls_145   -0.115     5
>  HD2    opls_146    0.115     5
>    C    opls_235    0.500     9
>    O    opls_236   -0.500     9
> 
> can that have a further effect since it seems that the file handling
> is based on line position and not atoms name.
> 

If you re-arrange the .rtp file, you'll have to start all over again, since only 
  pdb2gmx uses the .rtp files.  I doubt this is meaningful.  If you've got a 
.pdb file with HN as the amide proton, then it won't match the topology, which 
specifies this atom as H.  There will still be a name mismatch, but not an order 
mismatch, which simply indicates to me that you are using the wrong coordinate file.

-Justin

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================