[gmx-users] Problem implementing REST

[carlo martinotti]

I'm running GROMACS version 4.6.3, it seems that the hyperlinks doesn't work in the mailing list, so i'll post the links to my files (perturbed atoms are indicated by the "&"):
mdp > http://textuploader.com/oj7qtop > http://textuploader.com/oj7cffbonded > http://textuploader.com/oj7mffnonbonded >  http://textuploader.com/oj7h
The issue seems to be regargind lines in topology which correspond to proper dihedrals (function 9). Actually i've gone in the ffbonded file and i tried to delete the perturbed proper dihedrals, but then again it would give one error and one warning for each of the dihedrals as this one :

ERROR 48 [file topol2.top, line 222]:  Cannot automatically perturb a torsion with multiple terms to different  form.  Please specify perturbed parameters manually for this torsion in your  topology!

WARNING 48 [file topol2.top, line 222]:  No default Proper Dih. types for perturbed atoms, using normal values

The problem seems to be as it follows:
The author in the article claims to have made everything with the lambda system , but lambda system seems not able to perturb dihedrals with multiple terms, nor to use the same terms in state A and B. So, can i solve the problem and still use lambda system to interpolate betwen the 2 hamiltonians?
In case i can't, can i build one complete topology file for each replica that i want to simulate, perturb the topology modifying the hamiltonian consequently and then run a replica exchange  with the REMD routine? Wouldn't it be the same thing as using the lambda system, except it would be better because you can use verlet cutoff scheme, and so, also, GPUs?


> Date: Thu, 16 Oct 2014 12:04:38 +0200
> From: mark.j.abraham at gmail.com
> To: gmx-users at gromacs.org
> Subject: Re: [gmx-users] Problem implementing REST
> 
> Hi,
> 
> What GROMACS version is this? What bits of your topology are being
> complained about?
> 
> Mark
> 
> On Thu, Oct 16, 2014 at 11:12 AM, carlo martinotti <
> specialkender at hotmail.com> wrote:
> 
> > I'm currently trying to implement Replica Exchange with Solute Tempering
> > in gromacs, as previously done by terakawa et al (On easy implementation of
> > a variant of the replica exchange with solute tempering in GROMACS, J
> > Comput Chem. 2011 May;32(7):1228-34. doi: 10.1002/jcc.21703. Epub 2010 Nov
> > 29.)
> > From what i understood you have to change ffnonbonded.itp adding new
> > atomtypes with perturbed LJ parameters, then change ffbonded.itp adding new
> > bond/angles and dihedral types between this perturbed atoms with modifed
> > harmonic constant, and that's it for the forcefield.
> > Then you have to prepare a topology with state A and state B with state B
> > having perturbed charges and the atom types corresponding to the perturbed
> > version of the state A atoms.
> > Now, i have some problems. I'm trying to modify amber03 forcefield, but
> > when i try to run grompp it gives me this errors:
> >
> > ERROR 1 [file topol.top, line 199]:  Cannot automatically perturb a
> > torsion with multiple terms to different  form.  Please specify perturbed
> > parameters manually for this torsion in your  topology!
> > ERROR 2 [file topol.top, line 202]:  Cannot automatically perturb a
> > torsion with multiple terms to different  form.  Please specify perturbed
> > parameters manually for this torsion in your  topology!
> >
> > ERROR 3 [file topol.top, line 205]:  Cannot automatically perturb a
> > torsion with multiple terms to different  form.  Please specify perturbed
> > parameters manually for this torsion in your  topology!
> >
> > ERROR 4 [file topol.top, line 206]:  Cannot automatically perturb a
> > torsion with multiple terms to different  form.  Please specify perturbed
> > parameters manually for this torsion in your  topology!
> > Now, it seems that gromacs can't do this operation with amber, but in the
> > article that i read they actually used amber03, so what should i do? This
> > are my files mdp top ffbonded ffnonbonded
> > Can somebody explain me where is the problem?
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.