[gmx-users] parameters obtained after NPT

Justin Lemkul jalemkul at vt.edu
Thu Mar 26 22:54:46 CET 2015



On 3/26/15 12:33 AM, soumadwip ghosh wrote:
> Thanks Justin for a prompt reply.
>   Currently I am working with 4.5.6 and CHARMM 27 as I mentioned and I was
> wondering whether correct parameters can be obtained after NPT without
> really upgrading the software version because some of my previous results
> are obtained by working with 4.5.6 and CHARMM27 where I dint encounter such
> errors. When I view the trajectory using VMD the box looks deformed and the

Hopefully you haven't been using those plain 1-nm cutoffs.  Those are plainly wrong.

> water molecules are not well distributed. It might sound unrealistic but is
> there a possibility by which I can fix these errors in 4.5.6 and get the
> parameters right?
>
> Is it a problem that I am doing NPT first and then going for NVT. Ideally,
> it should'nt be a problem right?
>

No, that shouldn't be a problem.

> If my system is still blowing up why those nasty errors are not occurring
> this time? Do u suggest me to minimize the energy further? The potential
> got converged in the first attempt and thats why I was to equilibrate my
> system.
>

If the outcome is as striking as you make it sound, simply visualizing the 
trajectory should give some clear indication as to what has happened.

> How can I ensure that by upgrading the version and using correct mdp
> parameters I wont be facing such errors? I mean are systems such as mine

I have personally verified that energies and forces agree between CHARMM and 
GROMACS with those settings.  Other CHARMM labs have done the same.

> (CNT+DNA) known to have problems in equilibration in CHARMM27 (NPT with
> position restraints)?
>

CNT aren't explicitly included in CHARMM force fields, so your parameters there 
must be of your own creation or from some outside source.  DNA should be no 
problem (though CHARMM36 is newer and is favored over CHARMM27).

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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


More information about the gromacs.org_gmx-users mailing list