[gmx-users] md after digging a hole in the membrane

[Jason DeJoannis]

Erik,

 While you are on the subject of PME accuracy
formulas, what about eqn 32 of Petersen, JCP, 103,
3668, 1995? I have not seen much use of this 
in the literature. I have used accuracy formulas
for Ewald and P3M myself and they are quite convenient.

regards,

Jason

---
Jason de Joannis, Ph.D.
Chemistry Department, Emory University
1515 Pierce Dr. NE, Atlanta, GA 30322
Phone: (404) 712-2983
Email: jdejoan at emory.edu
http://userwww.service.emory.edu/~jdejoan



Quoting Erik Lindahl <lindahl at stanford.edu>:

> Hi,
> 
> I'm not quite sure which box you are talking about, since 6.18,6.18,6.66 
> are all larger than 0.9 :-)
> 
> Perhaps it is a simple formatting error in the coordinate file?
> 
> In any case, a system with box dimensions in the size of 2nm might 
> behave pretty strange with lipid molecules that themselves are 3 nm long.
> 
> By the way: rcoulomb isn't really a cutoff, since we do the rest in 
> reciprocal space. You can always decrease rcoulomb and decrease the 
> fourier grid spacing to do a more accurate sum in reciprocal space. 
> (That would actually be an incredibly useful diploma work if somebody 
> has an interested student - calculating a good set of accuracies for 
> different cutoff and grid spacing combinations).
> 
> Cheers,
> 
> Erik
> 
> Itamar Kass wrote:
> 
> >   Dear all, I try to use the DMPC bilayer from Tieleman's site in order to
> >produce a hole using Graham's modified mdrun.  The dimensions of the system
> (a
> >box of DMPC with water) are 6.176*6.176*6.6608 nm^3 (as given in the pdb
> file).
> > In order to relaxed the system after digging a hole in it, I wanted to do
> some
> >MD (before entering my protein). 
> >   One of the variables I entered in the md.mdp file is "rcoulomb= 0.9",
> were
> >the coulomb interaction type was set to PME.  The grompp program replays
> "ERROR:
> >One of the box vectors is shorter than twice the cut-off length. Increase
> the
> >box size or decrease rcoulomb.".  Yes I know this is true, no need to tell
> me
> >this, my questions are: 
> >(i) Should I lower the rcoulomb parameter, and also the rlist and rvdw to
> ~6.1
> >nm?  Will the validity/reliability of the later simulation (which will be
> given
> >the same parameters) is going to change, using smaller rcoulomb,alist and
> rvdw
> >lower then 0.9nm (which is already low)? 
> >(ii) Should I enlarge the system by adding DMPC molecule and then solvated
> then?
> > If yes, how can I do this easel?  I saw in the article of Faraldo-Gomez et.
> al.
> >(Eur. Biophys. J. 2002 Jun;31(3):217-27) that their systems were much
> bigger
> >then the one I use.  
> >(iii) Is there any place other then Tieleman's site where I can download
> DMPC
> >bilayer structures? 
> > 
> >       Thanks you all, Kass.
> >
> >
> >
> >********************************************
> >Computers are like airconditioners... They don't work well with Windows
> >open.
> >********************************************
> >
> >===========================================
> >| Itamar Kass
> >| The Alexander Silberman
> >| Institute of Life Sciences
> >| Department of Biological Chemistry
> >| The Hebrew University, Givat-Ram
> >| Jerusalem, 91904, Israel
> >| Tel: +972-(0)2-6585146
> >| Fax: +972-(0)2-6584329
> >| Email: ikass at cc.huji.ac.il
> >| Group Homepage: http://www.ls.huji.ac.il/~membranelab/
> >============================================
> >
> >_______________________________________________
> >gmx-users mailing list
> >gmx-users at gromacs.org
> >http://www.gromacs.org/mailman/listinfo/gmx-users
> >Please don't post (un)subscribe requests to the list. Use the 
> >www interface or send it to gmx-users-request at gromacs.org.
> >
> 
> 
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
>