From: Aron Broom (broomsday_at_gmail.com)
Date: Mon Apr 30 2012 - 01:40:35 CDT
If your protein really is roughly spherical, then your box should be
roughly cubic if you've simply added waters to a certain distance from your
protein. As Norman points out, a perfectly cubic box is hardly needed, but
I imagine it would be necessary if you had a very elongated protein to
ensure all box dimensions were at least longer than the longest axis of the
protein.
Anyway, the point is, if your box isn't fairly cubic, but your protein is
roughly spherical, I think you did something silly when solvating your
system.
~Aron
On Mon, Apr 30, 2012 at 1:29 AM, Norman Geist <
norman.geist_at_uni-greifswald.de> wrote:
> Xi,****
>
> ** **
>
> there’s is no need to fit a box shape to the proteins shape. The box is
> periodic so it doesn’t matter if part of the protein cross the boundaries.
> One can do such thing to save computing time, for that reason there’s also
> a spherical boundary condition (see manual) , as one don’t need to carry
> all those additional waters, that needs to be simulated when the box is too
> big for example. Also keep in mind that your box should be big enough, that
> the protein doesn’t interact with itself periodically, as this is mostly
> not wanted. So the bottom line is, that one don’t need a box that perfectly
> fits the solute, if one doesn’t have a molecule, that needs to be continued
> periodically like a metal surface or a membrane for example. I think you
> should inform about what periodic cell actually means.****
>
> ** **
>
> ** **
>
> Norman Geist.****
>
> ** **
>
> *Von:* Xi Cheng [mailto:xichengeva_at_gmail.com]
> *Gesendet:* Freitag, 27. April 2012 17:20
> *An:* Norman Geist
> *Betreff:* Re: AW: namd-l: Set up NPT simulation in cubic PBC-box****
>
> ** **
>
> Hi Norman,****
>
> ** **
>
> Thanks for your reply. ****
>
> ** **
>
> I want to simulate a sphere shape protein in the solution in NPT
> condition. ****
>
> I tried to fix the ratio between x and y, and start system with cubic PBC
> box. ****
>
> However, the resulting system has long box length along z; twice as those
> along x or y. ****
>
> ** **
>
> How can I keep the box cubic?****
>
> Or is there any other way that I can run NPT simulation for sphere shape
> protein?****
>
> ** **
>
> Thanks,****
>
> Xi****
>
> ** **
>
> On Apr 27, 2012, at 12:42 AM, Norman Geist wrote:****
>
>
>
> ****
>
> Hi Xi,****
>
> ****
>
> as far as I know, u can only fix the ratio between x and y. Maybe we can
> help you, when you explain what you want to do, and why you want to have an
> exactly cubic box.****
>
> ****
>
> Norman Geist.****
>
> ****
>
> *Von:* owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] *Im
> Auftrag von *Xi Cheng
> *Gesendet:* Donnerstag, 26. April 2012 23:23
> *An:* namd-l_at_ks.uiuc.edu
> *Betreff:* namd-l: Set up NPT simulation in cubic PBC-box****
>
> ****
>
> Hi, ****
>
> ****
>
> Is it possible to *allow the three orthogonal dimensions of the periodic
> cell to fluctuate* ****
>
> and *keep the ratio of the unit cell in x-y/y-z/z-x plane constant at the
> same time*?****
>
> ****
>
> Or is there any other way to *run NPT simulation in cubic PBC box*?****
>
> ****
>
> Thanks,****
>
> Xi****
>
> ****
>
> ****
>
> ** **
>
-- Aron Broom M.Sc PhD Student Department of Chemistry University of Waterloo
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:21:55 CST