VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Mar 30 2005 - 16:08:25 CST

Hi Luis, Nicolas,
  Regarding your questions about viewing virus structures:
Yes, I've rendered some huge virus structures with VMD. :-)
The main issue with this is that MSMS doesn't like calculating
surfaces for molecules with large holes in them (like virus structures)
so you have to convince MSMS to do it by making a separate MSMS
representation for each of the subunits. (irritating, I know...)
This problem has nothing to do with a machine limitation, this
is just how MSMS works. You should email Michel Sanner if
you have ideas or suggestions for getting MSMS to work better for
structures of this size.

We've got a variety of visualization machines in our lab here.
The main issue when working with huge structures is that you
have an adequate amount of RAM, and that you're running an OS
that actually lets you use it for VMD. All of our visualization
machines are now 64-bit processors running either Solaris or Linux
on them. (Solaris on the Sparc machines and Linux on the AMD Opterons)

Most of the desktops used by our researchers are now Sun W2100z Opteron
boxes with 4GB or 8GB of ram, running Red Hat Linux (RHEL3u4). Our
machines for sterescopic display are mostly Sun workstations with 4GB
of ram and one with 32GB. There's a list of the equipment we use here:
  http://www.ks.uiuc.edu/Development/Computers/systems.html

All of these machines easily cope with the 1-4 million atom system sizes
you mention, though the display update rate depends heavily on the
representations you're using etc.

  John Stone
  vmd_at_ks.uiuc.edu

On Wed, Mar 30, 2005 at 06:33:21PM -0500, Luis Rosales wrote:
>
> ---------- Original Message -----------
> From: Nicolas CHARVIN <Nicolas.Charvin_at_univ-savoie.fr>
> To: "Luis Rosales" <ludwig_at_biomedicas.unam.mx>
> Sent: Wed, 30 Mar 2005 13:59:43 +0200
> Subject: RE: vmd-l: Maximum numbers of atoms
>
> > >As far as I know, VMD dont have a limit on the number of atoms that can
> > >handle, (I think that is mentioned on the user manual
> >
> > yes, I'v read it that too.
> > But, on my Athlon XP3000+, with 2 GB of RAM, I was unable to read a
> > 640,000 atoms PDB file. It crashed.
>
> Hi,
>
> The machine is a SGI Onyx 350 with 12 processors MIPS R16000 64-bit @ 700 Mhz,
> 4MB caché. it has 3 graphic pipes with 2 RasterManager InfiniteReality4 each,
> 24 GB SDRAM.
>
> The manipulation of the molecule was very smooth, in the lines representation
> I was able to rotate easily the whole virus. I dont remember the time it took
> to calculate the MSMS surface, but at the end of the process it only showed
> the surface of just one of the 180 units composing the virus... :(
>
> However, before working on the Onyx, I was able to visualize and rotate the
> virus in line mode and CA tube on a workstation with 2 Xeon processors at 1.2
> GHz and 1 Gb of RAM the graphic card was a Nvidia Gforce with 32 Mb.
>
> By the way, John, have you visualized the surface on such huge structures???
> (I guess the answer is yes, as I saw the VMD spotlight on the virus, on the
> VMD homepage)
> I am still interested on the visualization of the surface of my virus and I
> would like to know why I am unable to do that (if its a hardware issue or a
> configuration issue)...
>
> >
> > >I have read 1,100,000 atoms (a virus...) on VMD, as a matter of fact I was
> > >able to generate the respective PSF file, etc...
> >
> > So, it is possible to load a 1 atom file....
>
> > Could you tell with precisely which hardware configuration you used,
> > and what you were able to do (wire reprensation ? VDW ? how smooth
> > was the rotation ? ). These kind of details wouls be very
> > appreciated, since we are on the way to buy a new visualization
> > workstation, and we hesitate whether a strong PC would be enough.
> >
> > Thanks in advance
> >
> > nicolas
> >
> > >However even on our SGI Onyx (24 Gb of RAM) I have been unable to display a
> > >good MSMS surface for my virus. :(
> > >
> > >Good luck,
> > >
> > >Luis
> > >
> > >
> > --
> > Nicolas.Charvin_at_univ-savoie.fr
> > LMOPS - Lab. Matériaux Organiques à Propriétés Spécifiques
> > Bât IUT - Université de Savoie
> > 73376 LE BOURGET DU LAC CEDEX
> > Tel: 04-79-75-86-53
> >
> > http://www.univ-savoie.fr/labos/lmops
> ------- End of Original Message ------- 

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349              
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078