[gmx-users] RE: FEP : separating components of dgdl

[Berk Hess]

----------------------------------------
> Date: Mon, 4 Feb 2008 16:18:53 +0100
> From: michael.bon at cea.fr
> To: gmx-users at gromacs.org
> Subject: FEP : separating components of dgdl 
> 
> 
> 
> Thanks Maik for your parameters for amber99 soft core !
> 
> Berk, you were right. The missing term is indeed the one associated to the use of constraints=all-bonds. And yes, there was a mistake in my .top file...
> 
> May I ask an other question ?
> 
> I still have a warning message with grompp :
> "WARNING 1 [file "XXX.itp", line 3451]:
>    Some parameters for bonded interaction involving perturbed atoms are specified
>    explicitly in state A, but not B - copying A to B"
> 
> where as the line 3451 is  (dihedral) :
> 
> 175 177 178 187  1 nucleic_imp_10  nucleic_imp_10
> 
> nucelic_imp_10 being defined in ffamber99bon.itp.
> 
> It seems to me that the state B is explicitly specified. How to get rid of this warning ? It does what I want anyway, but I want to understand what is happening.
> 
> Thanks,
> Michaël Bon

The problem here is that in ffamber99bon.itp nucleic_imp_10 is defined as:
#define nucleic_imp_10 180.00     4.60240      2    ; mis...
So what you get after cpp has processed this is:
175 177 178 187  1 180.00     4.60240      2    ; mis... 180.00     4.60240      2    ; mis...

Thus the B-state parameters are in the comment.
The solution is to move the comment in ffamber99bon.itp off the same line
as the parameters.

In Gromacs4.0 we could fix this by removing comments in definitions,
since there we will use our own preprocessing code.

Berk.

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/