[gmx-users] Re: grompp error- Excluding 3 bonded neighbours-Fatal (Dean Cuebas)

[Mark Abraham]

Dean Cuebas wrote:
>>> Dean Cuebas wrote:
>>> Dear users,
>>> I¹ve been trying to reproduce the GROMACS tutorial for Drug _ Enzyme Complex
>>>
>>> I¹ve tried to get grompp to work, but I keep getting the following message:
>>> _______________________________________________________
>>>
>>> creating statusfile for 1 node...
>>>
>>> Back Off! I just backed up mdout.mdp to ./#mdout.mdp.1#
>>> checking input for internal consistency...
>>> calling /usr/bin/cpp...
>>> processing topology...
>>> Generated 165 of the 1596 non-bonded parameter combinations
>>> Excluding 3 bonded neighbours for Protein             1
>>> Excluding 3 bonded neighbours for IN4                 1
>>> Excluding 2 bonded neighbours for SOL              8756
>>> NOTE:
>>>   System has non-zero total charge: 7.999998e+00
>>>
>>> processing coordinates...
>>>
>>> -------------------------------------------------------
>>> Program grompp, VERSION 3.3.2
>>> Source code file: grompp.c, line: 448
>>>
>>> Fatal error:
>>> number of coordinates in coordinate file (trp_b4ion.pdb, 28394)
>>>              does not match topology (trp.top, 28402)
>>> -------------------------------------------------------
>>>
>>> It appears because grompp is ³excluding² the bonded neighbors is what¹s
>>> making me end up with the different number of coordinates in my coordinate
>>> and topology files.
>>>
>>  No, "exclusions" are something else related to the structure of the
>>  force field - atoms connected by bonds shouldn't also be modeled with
>>  non-bonded interactions. The coincidence of 28402-28394=8 equaling 3+3+2
>>  is just that.
>>  
>>  Rather more significant is that you have a non-zero charge of (rounded)
>>  8 and you seem to be part way through a process of replacing water
>>  molecules with ions, which usually leaves a system neutral. I'd guess  you've
>> done something wrong there - most likely mislabeled a file.
>>  
>>  Mark
> 
> Thanks for your comment, I know realize the issue has nothing to do with
> "exclusions but I still don't understand the problem I have.
> 
> I have the same # of molecules in the .top and the .pdf files that are going
> into grompp: 1 protein, 1 ligand, and 8756 SOL. Everywhere I read, it says
> that you will get a "warning" noting that the system has a nonzero charge,
> before you've added any ions. That I understand, and people still generate
> .tpr files to then use in genion to fix the charges.
> But, why could the log say I have  8 more sets of coordinates in my .top
> file than my .pdb file, which causes a fatal error, when I have identical #
> of molecules????

Atoms != molecules. In the process of adding ions to neutralize, genion 
will remove 3-atom water molecule(s) and replace them with single-atom 
ions, leading to preservation of the number of molecules, but changes in 
the number of atoms. The [ molecules ] section of the topology will need 
to be updated consistently, as the tutorial will be telling you. If you 
get some part of the above wrong, you'll observe symptoms such as you 
are observing. There are other hypotheses... but going back to the start 
of the tutorial and trying again taking more care is most likely what 
you need to do. :-)

> In fact, I tested my system with the gmxtest suite 3.3.2, and got the
> following simple error which also claims that a .tpr file can't be made:
> 
> No topol.tpr file in morse. grompp failed
> FAILED. Check files in morse
> 1 out of 16 simple tests FAILED
> FAILED. Check files in acetonitrilRF
> FAILED. Check files in dec+water
> 2 out of 14 complex tests FAILED
> All 63 kernel tests PASSED
> All 45 pdb2gmx tests PASSED

If I recall correctly, the test suite was designed for 3.3.3, and isn't 
really robust enough to be used as a routine diagnostic tool. Search the 
list archives for more info if you like.

> (I also reinstalled gromacs and I get the same problem)
> 
> I've followed the tutorial steps, except that I'm using version 3.3.2 on
> MacOS X 10.5.6.  Any help would be so greatly appreciated.
> 
> I plan to "update that tutorial for gromacs 3.3.2" and post it back to the
> gromacs site soon... IF I can get past this point.

Good luck!

Mark