Re: AW: Energy minimization of membrane

From: R. Charbel Maroun (charbel.maroun_at_inserm.fr)
Date: Fri Feb 03 2012 - 04:58:17 CST

Hi,

Thanks for the feedback. It's not that I didn't want to use PBC. I
just wanted to know if it was necessary for a minimization. By the
way, before running MD, the system must be well minimized, otherwise
it might become unstable during the trajectory. It's part of the
equilibration step, before production.

I used CHARMM-GUI for building my membrane system. I ran last night an
energy minimization with PME and PBC for just the
water molecules and the K and Cl ions; lipids and receptor fixed (when
you use PME you need PBC). The question is what values to use for
PMEGridSize and cellBasisVector? What I did is that for the first set
I measured approximately on the screen the dimensions of the box size;
for the second set I just guessed slightly larger values. I don't know
if that's right.

The results of the minimization show that the ordered structure of
waters away from the phospholipid head groups was lost (good!). The
water layers above and below got flattened, more compact along the
axis perpendicular to the membrane, and wider in the axes of the
membrane. When I ran a minimization without PBC, one water molecule
would always escape from the box and get wild coordinates. NAMD would
then stop if you tried to use this coordinate set for further
calculations.

# PBC
cellBasisVector1 90.0 0.0 0.0
cellBasisVector2 0.0 109.0 0.0
cellBasisVector3 0.0 0.0 90.0
cellOrigin 0.0 0.0 2.0
wrapAll on
wrapWater on
dcdUnitCell yes

# PME
PME yes
PMEGridSizeX 85
PMEGridSizeY 85
PMEGridSizeZ 103

Any suggestions ?

Cheers,

C Maroun

Norman Geist <norman.geist_at_uni-greifswald.de> a écrit :

> Hi,
>
>
>
> I'm not a specialist in bio chemistry, more on IT, but I would expect your
> membrane as not big enough to be realistic without its periodic copys. So it
> won't behave good, as it would be in vacuum, and the best energy
> conformation achieved then, wouldn't be the one you want. I would expect
> that it will relax more into a strange conformation, as in a realistic one.
> Also, if you don't set PBC, your solvent will diffuse away, if there is one.
>
>
>
> What's the reason for not want to set PBC in minimization? If you want to
> look if your system want to expand, you can try that out with some different
> box sizes, and with dynamics, not minimization as minimization will not move
> the system much.
>
>
>
> Correct me if required ;)
>
>
>
> Best wishes
>
>
>
> Norman Geist.
>
>
>
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag
> von Boyang Wang
> Gesendet: Freitag, 3. Februar 2012 06:31
> An: R. Charbel Maroun
> Cc: namd-l_at_ks.uiuc.edu
> Betreff: Re: namd-l: Energy minimization of membrane
>
>
>
> Hi Charbel,
>
>
>
> I think PBC is necessary, but keep in mind that calculations will be much
> slower.
>
>
>
> Best,
>
> Boyang
>
>
>
> On Thu, Feb 2, 2012 at 8:37 PM, R. Charbel Maroun <charbel.maroun_at_inserm.fr>
> wrote:
>
> Hello,
>
> When we do energy minimization with NAMD of a protein-membrane system,
> do we have to give PBC in the script, just like for a
> molecular dynamics calculation or is this implicitely taken into
> consideration? If not, that means that the protein-membrane system is
> in vacuum and could become unstable during minimization.
>
> Greetings,
>
> C Maroun
>
>
>
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
>
>
>
>
>
>
>
>
> --
> Boyang Wang, Ph.D.
>
>

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:24:42 CST