• ## Outreach

From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Jun 30 2004 - 13:04:29 CDT

Hi Randall,
I'm assuming you're working with higher resolution grids than
just 2 and 3 samples, but keep in mind that the isosurface extraction
algorithm in VMD will only generate geometry between the grid points,
not ON the grid points.. So, if you have a grid of resolution NxNxN,
the resulting surface would contain (N-1)x(N-1)x(N-1) "cells" which
each subsequently contain up to 6 triangles (with the current implementation).

If you're working with low resolution grids, it is easy to get assymmetric
surfaces particularly if working with an even grid resolution and
surface shapes that are figures of revolution.

All this being said, it might be easier to debug your problem if you
send a copy of the data file so we can take a look at it here.
I won't have a chance to look at it until next week as I'm out of town
for an NIH advisory board meeting the rest of this week, but I can look
at it shortly after I return.

Thanks,
John Stone
vmd_at_ks.uiuc.edu

On Wed, Jun 30, 2004 at 10:58:58AM -0500, Randall Hall wrote:
> Hi John, thanks for your reply. I am using VMD 1.8.2 (for mac) and
> the pltfile program i got from from the opengmol site (i think that is
> where it is located). the display i described was for an even number
> of points. if i use and odd number of points, the sphere is symmetric,
> but not centered at the origin (i am reading in a pdb file with a few
> atoms, but they are already centered at the origin), but 1/2 way
> between the origin and the next grid point. an example may help
> explain. (1-D for simplicity)
>
> assume a box of side length L, centered at the origin. Put in atoms,
> with coordinates (-L/2, 0, and +L/2). If i use 2 grid points, I
> evaluate the spherical density at -L/2 and 0. The display would then
> have the correct value at -L/2 and 0, but nothing gets displayed at
> +L/2. If I use 3 grid points, I evaluate the spherical density at
> -L/2, 0, and +L/2. The displayed spherical density then has its
> maximum displayed at -L/4. I think this may be a simple problem of how
> VMD decides where to plot things.
>
> Randy Hall
>
>
> Randall W. Hall
> Professor of Chemistry
> Louisiana State University
> Baton Rouge, LA 70803-1804
> Email: rhall_at_lsu.edu Web: http://chemistry.lsu.edu/hall
> Phone: 225-578-3472 Fax: 225-578-3458
> On Jun 28, 2004, at 3:11 PM, John Stone wrote:
>
>
> >
> > Dear Randall,
> > Unless you've chosen a grid resolution with an odd number of
> > sample
> > points, and the grid is precisely centered, the result you mention
> > would
> > not be surprising, since the grid points and thus the borders of
> > the
> > volume would not be concentric with the sphere you generated
> > density
> > values from. I assume you're using VMD 1.8.2? Are you using the
> > latest
> > version of the plugins, or an older version? (we fixed a bug in
> > the PLT
> > plugin a while back, as I recall)
> >
> > Thanks,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Mon, Jun 28, 2004 at 10:14:20AM -0500, Randall Hall wrote:
> >
> > > Hello, I have a question about how VMD is displaying an
> > > isosurface I
> > > have created in .plt format. As a test, I created a gaussian
> > > density
> > > centered at the origin of the box. When the isosurface is
> > > displayed, I
> > > see a sphere for large isovalues, as I expect. When I reduce the
> > >
> > > isovalue, the sphere grows in size until the sides bump into the
> > > walls
> > > of the box and "holes" develop in the sphere. What is surprising
> > > to me
> > > is that the holes do not appear symmetrically. It appears that
> > > the
> > > isosurface is truncated before it hits the wall in positive x, y,
> > > and z
> > > directions. That is, the sphere never reaches 3 of the size box
> > > sides.
> > > Can anyone explain why this truncation is happening? thanks
> > >
> > > Randall W. Hall
> > > Professor of Chemistry
> > > Louisiana State University
> > > Baton Rouge, LA 70803-1804
> > > Email: rhall_at_lsu.edu Web: http://chemistry.lsu.edu/hall
> > > Phone: 225-578-3472 Fax: 225-578-3458
> > >
> > >
> >
> > --
> > 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
> >
>
>

```--
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
```