From: Richard Swenson (swenson_at_hec.utah.edu)
Date: Thu Nov 01 2007 - 11:53:28 CDT
Oops, I forgot to CC the NAMD list!
Sent: Thursday, November 01, 2007 10:51 AM
To: 'Himanshu Khandelia'
Subject: RE: namd-l: Controlling the area per lipid
There are those in the CHARMM community that use a constant surface tension
(NPzGammaT) to produce bilayers with the correct area per lipid. However,
there is debate about what this imposed tension to a flat bilayer means
physically. We currently use NPT dynamics to allow thermal bilayer
undulations and other membrane deformations in our large systems.
A quick google search yielded this paper that may be of some interest to
Simulation-Based Methods for Interpreting X-Ray Data from Lipid Bilayers
Biophysical Journal, Apr 15, 2006, by Klauda, Jeffery B, Kucerka, Norbert,
Brooks, Bernard R, Pastor, Richard W, Nagle, John F
From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf
Of Himanshu Khandelia
Sent: Thursday, November 01, 2007 5:20 AM
To: NAMD list
Subject: Re: namd-l: Controlling the area per lipid
Please be warned that if you use the CHARMM force field and the NPT
ensemble, the membrane will NOT equilibrate to the correct area per lipid,
the area will keep shrinking. The NPT ensemble is not best suited to
simulate a membrane with the CHARMM force field.
A good strategy might be to start with dimensions 10% larger than the
known experimental area per lipid, and allow the area to shrink (in NPT)
to the correct value area per lipid, and then fix the area of the bilayer
and continue your simulation in the NPzAT ensemble.
Research Assistant Professor (Postdoc),
MEMPHYS, Center for BioMembrane Physics: www.memphys.sdu.dk
University of Southern Denmark (SDU)
Campusvej 55, Odense M 5230, Denmark
Phone: +45 6550 3510, +45 2398 7972
Fax: +45 6550 4048
On Wed, 31 Oct 2007, Ilya Chorny wrote:
> Should I just use constant pressure to allow the membrane to shrink. What
> the lateral pressure of a membrane, 1 ATM? What if I just decrease the X
> and Y dimensions?
> On 10/31/07, JC Gumbart <gumbart_at_ks.uiuc.edu> wrote:
> > Minmax measurements of lipids will not be reliable. Lipids that
> > stick out of the simulation cell will still be measured because they
> > don't get wrapped until most of the molecule crosses the periodic
> > boundary. The water will be more accurate, but the numbers in the
> > xst file will be the exact cell size over the simulation. If you are
> > using constant area, the x-y numbers should never change.
> > You should consider during initial equilibration though that the
> > membrane may shrink (and may need to), especially if you put a
> > protein in the middle and removed all lipids near it. After that,
> > you could run constant area though.
> > On Oct 31, 2007, at 8:23 PM, Ilya Chorny wrote:
> > > Hi I am having some trouble controlling the area per lipid molecule
> > > in my simulation. I set my cell basis vector in the XY plane to the
> > > correct area to maintain the proper density but the minmax
> > > measurements of my lipid give me values that are larger than the
> > > cell basis vectors. I also you constant area pressure coupling.
> > > Why is that? Any ideas on how I can maintain the correct density?
> > >
> > > Thanks,
> > >
> > > Ilya
> > >
> > >
> > > --
> > > Ilya Chorny Ph.D.
> Ilya Chorny Ph.D.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:27 CST