[gmx-users] genion

Justin Lemkul jalemkul at vt.edu
Mon Aug 15 13:46:35 CEST 2016

On 8/15/16 4:11 AM, f.namazifar at sci.ui.ac.ir wrote:
> Hi every body;
> Why we should neutralise net charge of the simulation system?

To properly answer your question, we must approach it with a bit more nuance.

One typically neutralizes the net charge of a condensed-phase system for two 
reasons, one physical and one algorithmic.  First, in e.g. aqueous solution, 
positives don't exist without negatives.  That's just fundamental physics.  If a 
species ionizes in solution, its corresponding counterion is generated.  Second, 
the de facto standard for computing electrostatic interactions in an MD 
simulation using PBC is PME, which requires a net-neutral simulation system. 
One does *not* have to add counterions to achieve net neutrality, as the PME 
algorithm itself contibutes a uniform background plasma to neutralize the 
charge.  The problem is that, in heterogeneous systems, this uniform background 
charge leads to artifactual behavior.  Most all biomolecular simulations are 
heterogeneous, so it is unwise to rely on PME itself to account for this effect. 
  So the very simple solution is to add counterions.  Generally, one should add 
some additional ions to mimic in vitro or in vivo conditions (e.g. for 
biomolecules) but the exact approach depends on what is being studied.

There are other systems, for instance an ionized protein in vacuo, for which you 
would not add counterions (there's nowhere to add them, really).  But in these 
cases, you wouldn't be using PME, anyway.

So the real answer depends on what you are doing.  It is commonly accepted that 
with PME, the easiest thing to do is add a couple of ions to balance out the charge.



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


More information about the gromacs.org_gmx-users mailing list