[gmx-users] Values for pairs, when attached to the topology...?

Berk Hess gmx3 at hotmail.com
Fri May 13 15:26:57 CEST 2005

>From: Maik Goette <mgoette at mpi-bpc.mpg.de>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: GMX-mailinglist <gmx-users at gromacs.org>
>Subject: [gmx-users] Values for pairs, when attached to the topology...?
>Date: Fri, 13 May 2005 14:49:01 +0200
>Hi all
>I have a problem with parameters, manually included to the pairs section. 
>For FEP purposes, I have(?) to include the pair parameters for the 
>perturbed atoms ((A and B value) into the topology file (OPLSAA FF).
>I took the values eps and sig for the two atoms i and j from the nonbonded 
>topology, calculated the ij value via the formulas in the manual and 
>inserted these values for eps and sig in the topology file.
>mdrun claims, that the parameters are wrong:
>Fatal error: Force field inconsistency: 1-4 interaction parameters for 
>atoms 1615-1621 not the same as for other atoms with the same atom type
>Obviously, the parameters, I calculated are wrong. Do I have to include 
>fudge-factors in my calculations ?
>I am also not sure, if there have to be c6 and c12 in the topology file, or 
>if eps and sig are ok.
>Please give me some information, how the parameters out of the forcefield 
>have to be calculated correctly and which ones I have to use in my topology 

You only need to enter the B-topology parameters when they are
different from the A values.

Unfortunately there are two problems in 3.2.1:
the pair parameters should always be in c6/c12 form
and you can not have different pair parameters for the same
atom type combinations.
Both problems have been fixed in the CVS, but the CVS code
currently has no free-energy loops.

But I think that if you only change the charges and/or atom types
all bonded parameters (including the pairs) should be set
correctly for the B topology by grompp.


Talk with your online friends with MSN Messenger http://messenger.msn.nl/

More information about the gromacs.org_gmx-users mailing list