pymol-users Mailing List for PyMOL Molecular Graphics System

Andreas,

Thanks for compiling & testing the current open-source code!  I think this
is a driver issue.  Some card/driver combinations object to being fed
geometry too fast in immediate mode.  Please See if one or both of the
following settings helps:

set ati_bugs
set nvidia_bugs

Also, try:

set use_display_lists

Cheers,
Warren

--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...
________________________________

From: pym...@li...
[mailto:pym...@li...] On Behalf Of Andreas
Forster
Sent: Saturday, May 31, 2008 3:32 AM
To: PYMOLBB
Subject: [PyMOL] exploding proteins


Dear all,

I just checked out PyMOL 1.1r0 revision 3367 via subversion.  When I open
any old pdb file (eg. fetch 1ubq), the protein is displayed as if irradiated
with a free-electron laser
(https://fileexchange.imperial.ac.uk/files/5b706c5efa/lines.png).  The
cartoon (https://fileexchange.imperial.ac.uk/files/0dd1d6af73/cartoon.png)
or any other representation is fine, except ribbon, which displays nothing.

Is that a bug in PyMOL or in my graphics driver?  I'm running Fedora 7 with
the current ATI driver obtained from the livna repository.  The graphics
card is a dusty FireGL T2.

Thanks.


Andreas

Dear all,

I just checked out PyMOL 1.1r0 revision 3367 via subversion.  When I open
any old pdb file (eg. fetch 1ubq), the protein is displayed as if irradiated
with a free-electron laser (
https://fileexchange.imperial.ac.uk/files/5b706c5efa/lines.png).  The
cartoon (https://fileexchange.imperial.ac.uk/files/0dd1d6af73/cartoon.png)
or any other representation is fine, except ribbon, which displays nothing.

Is that a bug in PyMOL or in my graphics driver?  I'm running Fedora 7 with
the current ATI driver obtained from the livna repository.  The graphics
card is a dusty FireGL T2.

Thanks.


Andreas

Frank,

Honestly, this is the job of the underlying graphics driver, and I suspect
that nVidia will soon support interleaved stereo 3D via the standard OpenGL
quadbuffer API.  They've shown a willingness to support quite a number of
different approaches already.

However, I amm wary of checkerboard-interleaved stereo 3D.  None of the
interleaved solutions I've seen to date can handle the kinds of thin-line
visualizations crystallographers typically employ nearly as well as
framebuffer-interleaved full-resolution 3D used with CRTs and active-stereo
DLP projectors (such as the Christie Mirage).

Thus, my ultimate hope is that DLP-based displays will eventually bypass all
of this checkerboard nonsense and simply support full-resolution active
stereo 3D (ideal) or passive dual-display-input passive stereo 3D (the next
best option).  Given that DLP projectors already do this, there seems to be
no fundamental technical reason why desktop DLP displays shouldn't be able
to do so as well.  

Cheers,
Warren

--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...


________________________________

From: pym...@li...
[mailto:pym...@li...] On Behalf Of S. Frank Yan
Sent: Friday, May 30, 2008 4:01 PM
To: pym...@li...
Subject: [PyMOL] Stereo output checkerboard format



Hi,

 

I was wondering if PyMOL will support so-called checkerboard stereo format
output.  This format is natively supported by the new DLP technology.  We
can finally get rid of the CRT monitors and put on the flat monitor and/or
projection for the general audience, e.g. medicinal chemists.  I think this
would be a huge boost for PyMOL.

 

Thanks,

Frank

 

 

Mac Users,

If you use Stereo 3D on Mac OS X and are happily running Tiger, please do
not upgrade to Leopard yet.  It seems that stereo 3D visualization is still
broken under Leopard 10.5.3. 

If you are presently considering purchase of a Mac Workstation for Stereo 3D
Graphics visualization, then you may wish to defer your purchase for the
time being or go with Linux unless you can confirm for certain that Tiger
will indeed run on the machine you are buying, since you can't necessarily
expect older operating systems to install on newer hardware.

I have reconfirmed with my developer relations at Apple that they are aware
of the severity of the problem and have the issue appropriately prioritized.
Unfortunately, there is just a bit too much good stuff going on with Apple
these days... 

Cheers,
Warren
--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...

Hi,

 

I was wondering if PyMOL will support so-called checkerboard stereo
format output.  This format is natively supported by the new DLP
technology.  We can finally get rid of the CRT monitors and put on the
flat monitor and/or projection for the general audience, e.g. medicinal
chemists.  I think this would be a huge boost for PyMOL.

 

Thanks,

Frank

 

 

One pure PyMOL-based approach is to use the surface_carve settings as
follows:

load $TUT/1hpv.pdb, tmp
extract lig, organic
extract prot, polymer
delete tmp

set surface_carve_cutoff, 4.5
set surface_carve_selection, lig
set surface_carve_normal_cutoff, -0.1

show surface, prot within 8 of lig
set two_sided_lighting
set transparency, 0.5
show sticks, lig
orient lig

This approach culls away any triangle more than 4.5 angstroms away from all
ligand atoms and any triangle not facing at least one ligand atom.  It tends
to produce a fairly clean pocket representation.

Also worth trying (after the above):

set surface_color, white
set surface_type, 2
unset ray_shadows

Cheers,
Warren

--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...
 
________________________________

From: pym...@li...
[mailto:pym...@li...] On Behalf Of Thomas S.
Leyh, Ph. D.
Sent: Tuesday, May 27, 2008 1:53 PM
To: pym...@li...
Subject: [PyMOL] interior protein-surfaces


     From time to time I find it valuable to view the suface of a
ligand-binding pocket looking out from the surface's interior.   This is
accomplished in a variety of ways - principally, zooming and clipping.  It
would be wonderful to be able to represent just the "skin" of the active
site wrapping around the ligands, like a sac - this seems quite difficult to
achieve.  A related problem is that ray-ing a surface that has been clipped
yields a collection of odd surface regions that represent poorly the
non-ray-ed version - it is as if the resolution of the ray it too gross to
capture the clipping edges. 
 
   Advice appreciated.
 
    Thanks,
 
    Tom Leyh

Warren,
    Thanks very much for your suggestions.  Turning on Millions of 
Colors seems to have solved the problem.

David

DeLano Scientific wrote:
> David,
> 
> Are the misclicks plausible or absurd?  For example, if you click on the
> left side of the screen are unrelated atoms being selected on the right
> side?
> 
> If so, then the problem can probably be corrected by running under Millions
> of Colors instead of Thousands (see Display under System Preferences).
> 
> If not, then try to rotate the camera and adjust the clipping planes so that
> no other atoms are displayed near the atoms you wish to click on.
> 
> Cheers,
> Warren
> 
> --
> DeLano Scientific LLC
> Subscriber Support Services
> mailto:su...@de...
> 
> 
> -----Original Message-----
> From: pym...@li...
> [mailto:pym...@li...] On Behalf Of David
> Goldenberg
> Sent: Thursday, May 29, 2008 10:49 AM
> To: pym...@li...
> Subject: [PyMOL] problem picking atoms
> 
> Hi,
>     I am using MacPyMol under OS X 10.4 on a G5 PowerMac, and I frequently
> have problems select atoms by clicking on them.  Very often, no matter how
> carefully I aim with the mouse, the atoms selected are far from the ones I
> wanted.  It seems that this should be simple!  Is there something I am
> missing?
> 
> Thanks,
> David Goldenberg
> --
> Department of Biology
> University of Utah
> 257 South 1400 East
> Salt Lake City, UT  84112-0840
> 
> Telephone: (801) 581-3885
> Fax:  (801) 581-2174
> 
> E-mail: gol...@bi...
> Lab web page: http://www.biology.utah.edu/goldenberg
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft Defy all challenges.
> Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 
> 

-- 
Department of Biology
University of Utah
257 South 1400 East
Salt Lake City, UT  84112-0840

Telephone: (801) 581-3885
Fax:  (801) 581-2174

E-mail: gol...@bi...
Lab web page: http://www.biology.utah.edu/goldenberg

David,

Are the misclicks plausible or absurd?  For example, if you click on the
left side of the screen are unrelated atoms being selected on the right
side?

If so, then the problem can probably be corrected by running under Millions
of Colors instead of Thousands (see Display under System Preferences).

If not, then try to rotate the camera and adjust the clipping planes so that
no other atoms are displayed near the atoms you wish to click on.

Cheers,
Warren

--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...


-----Original Message-----
From: pym...@li...
[mailto:pym...@li...] On Behalf Of David
Goldenberg
Sent: Thursday, May 29, 2008 10:49 AM
To: pym...@li...
Subject: [PyMOL] problem picking atoms

Hi,
    I am using MacPyMol under OS X 10.4 on a G5 PowerMac, and I frequently
have problems select atoms by clicking on them.  Very often, no matter how
carefully I aim with the mouse, the atoms selected are far from the ones I
wanted.  It seems that this should be simple!  Is there something I am
missing?

Thanks,
David Goldenberg
--
Department of Biology
University of Utah
257 South 1400 East
Salt Lake City, UT  84112-0840

Telephone: (801) 581-3885
Fax:  (801) 581-2174

E-mail: gol...@bi...
Lab web page: http://www.biology.utah.edu/goldenberg

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges.
Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
PyMOL-users mailing list
PyM...@li...
https://lists.sourceforge.net/lists/listinfo/pymol-users

Hi,
    I am using MacPyMol under OS X 10.4 on a G5 PowerMac, and I 
frequently have problems select atoms by clicking on them.  Very often, 
no matter how carefully I aim with the mouse, the atoms selected are far 
from the ones I wanted.  It seems that this should be simple!  Is there 
something I am missing?

Thanks,
David Goldenberg
-- 
Department of Biology
University of Utah
257 South 1400 East
Salt Lake City, UT  84112-0840

Telephone: (801) 581-3885
Fax:  (801) 581-2174

E-mail: gol...@bi...
Lab web page: http://www.biology.utah.edu/goldenberg

I'm not sure it's exactly what you're looking for, but this seems similar:

http://hollow.sourceforge.net/

On Tue, May 27, 2008 at 4:53 PM, Thomas S. Leyh, Ph. D. <le...@ae...>
wrote:

>       From time to time I find it valuable to view the suface of a
> ligand-binding pocket looking out from the surface's interior.   This is
> accomplished in a variety of ways - principally, zooming and clipping.  It
> would be wonderful to be able to represent just the "skin" of the active
> site wrapping around the ligands, like a sac - this seems quite difficult to
> achieve.  A related problem is that ray-ing a surface that has been clipped
> yields a collection of odd surface regions that represent poorly the
> non-ray-ed version - it is as if the resolution of the ray it too gross to
> capture the clipping edges.
>
>    Advice appreciated.
>
>     Thanks,
>
>     Tom Leyh
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>
>


-- 
Michael Lerner, Ph.D.
IRTA Postdoctoral Fellow
Laboratory of Computational Biology NIH/NHLBI
5635 Fishers Lane, Room T909
http://www.umich.edu/~mlerner

Hi Tom,

I'm not sure this (http://lmc.uab.cat/~xavi/) is what you're looking  
for.

If it is, you can build the surface in the CASTp server (or using the  
PyMOL plugin) and then display the surface of the pocket (the .poc  
object) of the residues forming the binding pocket (e.g. 5 Ang. around  
the ligand).

Hope it helps,

Xavier

Xavier Deupi
Laboratory of Computational Medicine
Biostatistics Unit. School of Medicine
Universitat Autonoma de Barcelona
Bellaterra, 08193 (Barcelona)
Catalunya, EU

Hi Martin,

Now why didn't I ever think of that? So simple... In the .pymolrc you can add:

import os
/def gro( fnm ):\
    os.system("editconf -f %s -o /tmp/tmp.pdb" % fnm )\
    cmd.load( "/tmp/tmp.pdb", fnm[:-4] )
cmd.extend( "gro", gro )

And it will do what you want...

Cheers,

Tsjerk

On Wed, May 28, 2008 at 11:21 AM, Martin Höfling <mar...@gm...> wrote:
> Hi all,
>
> here's a small script that allows loading of a gromacs index file into pymol.
> The corresponding molecule has to be loaded with preserved atom order.
>
> I have also a question/feature request concerning the "load" operation: Is it
> possible to register e.g. this script, to handle loading of .ndx files. This
> would in particular be nice to load .gro files with a small script calling
> editconf -f toload.gro -o /tmp/loadme.pdb and then loading the pdb.
>
> Did I miss that option or doesn't it exist (at the moment)?
>
> Best
>        Martin
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>
>



-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623

Hi all,

here's a small script that allows loading of a gromacs index file into pymol. 
The corresponding molecule has to be loaded with preserved atom order.

I have also a question/feature request concerning the "load" operation: Is it 
possible to register e.g. this script, to handle loading of .ndx files. This 
would in particular be nice to load .gro files with a small script calling 
editconf -f toload.gro -o /tmp/loadme.pdb and then loading the pdb.

Did I miss that option or doesn't it exist (at the moment)?

Best
	Martin

Hi Tom,

On Tue, May 27, 2008 at 10:53 PM, Thomas S. Leyh, Ph. D.
<le...@ae...> wrote:
>      From time to time I find it valuable to view the suface of a
> ligand-binding pocket looking out from the surface's interior.   This is
> accomplished in a variety of ways - principally, zooming and clipping.  It
> would be wonderful to be able to represent just the "skin" of the active
> site wrapping around the ligands, like a sac - this seems quite difficult to
> achieve.

Do you mean like what you would get from:

create protein, not ligands
hide everything
show sticks, ligands
show surface, protein within 5 of ligands

(using appropriate selections)?

Otherwise, could you give an example of what you mean (and what you
don't mean), e.g. a link to an image?

> A related problem is that ray-ing a surface that has been clipped
> yields a collection of odd surface regions that represent poorly the
> non-ray-ed version - it is as if the resolution of the ray it too gross to
> capture the clipping edges.
>

I'm not quite certain what I have to imagine here. Linking an image
from before raytracing and one after would be helpful.

>    Advice appreciated.
>
>     Thanks,
>
>     Tom Leyh

Cheers,

Tsjerk

-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623

Dear Martin

If you wish to add names at the top of each record, a short python program
is available in
http://www.redbrick.dcu.ie/~noel/talks/PythonForChemistry.pdf.

hope this helps

Tomoko


2008/5/24 Martin Stoermer <m.s...@im...>:

> Hi folks,
> we often use Pymol to visualize SD files overlayed on protein structures
> usually coming out of docking and other VS runs. I've noticed that when
> using the seq_view window, some SD files which have the names of individual
> ligands at the very top of each record in the SDF have these names usefully
> displayed in seq_view, but if the names are in a <name> field, they don't
> displayed, we just get '' '' type entries for each ligand. Whilst I
> understand that seq-view is primarily for looking at and manipulating
> portions of the protein sequence, is there some way to view the name fields
> in Pymol?
> e.g. the following type of entry would have ZINC09999999_0 displayed in the
> seq_view window:
>
> $$$$
> ZINC09999999_0
>   -OEChem-05230802443D
>
>  50 54  0     0  0  0  0  0  0999 V2000
>    40.7619   54.6488    8.1176 C   0  0  0  0  0  0  0  0  0  0  0  0
>    40.1173   55.8601    8.2935 C   0  0  0  0  0  0  0  0  0  0  0  0
>    41.2285   54.2825    6.8679 C   0  0  0  0  0  0  0  0  0  0  0  0
>
> but the following would just display " or sometimes UNK
>
>
> csCF1100/05020817022D
>
>   9  8  0  0  0  0  0  0  0  0999 V2000
>    10.9319  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
>    12.2508  -19.9323    0.0000 C   0  0  3  0  0  0  0  0  0  0  0  0
>    13.5698  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
>    14.8887  -19.9323    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
>    16.2077  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
>    17.5266  -19.9323    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
>    10.9319  -17.6478    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0
>     9.6129  -19.9323    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0
>    12.2508  -21.4553    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0
>   1  2  1  0        0
>   2  3  1  0        0
>   3  4  1  0        0
>   4  5  1  0        0
>   5  6  1  0        0
>   1  7  2  0        0
>   1  8  1  0        0
>   2  9  1  0        0
> M  END
> >  <Name> (1)
> (+/-)2-BROMOHEXANOIC ACID  (2-BROMOCAPROIC ACID)
>
> >  <Amount> (1)
> 25G
>
> >  <Supplier catalogue number> (1)
> 24,283-7
>
> >  <Location> (1)
> STORE
>
> >  <CAS_Number> (1)
> 2681-83-6
>
> >  <MW> (1)
> 195.0449
>
> $$$$
>
>
> thanks,
> Martin
>
>  Dr. Martin Stoermer
> Institute For Molecular Bioscience
> University of Queensland
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>
>

Hi folks,

we often use Pymol to visualize SD files overlayed on protein  
structures usually coming out of docking and other VS runs. I've  
noticed that when using the seq_view window, some SD files which have  
the names of individual ligands at the very top of each record in the  
SDF have these names usefully displayed in seq_view, but if the names  
are in a <name> field, they don't displayed, we just get '' '' type  
entries for each ligand. Whilst I understand that seq-view is  
primarily for looking at and manipulating portions of the protein  
sequence, is there some way to view the name fields in Pymol?
e.g. the following type of entry would have ZINC09999999_0 displayed  
in the seq_view window:

$$$$
ZINC09999999_0
   -OEChem-05230802443D

  50 54  0     0  0  0  0  0  0999 V2000
    40.7619   54.6488    8.1176 C   0  0  0  0  0  0  0  0  0  0  0  0
    40.1173   55.8601    8.2935 C   0  0  0  0  0  0  0  0  0  0  0  0
    41.2285   54.2825    6.8679 C   0  0  0  0  0  0  0  0  0  0  0  0

but the following would just display " or sometimes UNK


csCF1100/05020817022D

   9  8  0  0  0  0  0  0  0  0999 V2000
    10.9319  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    12.2508  -19.9323    0.0000 C   0  0  3  0  0  0  0  0  0  0  0  0
    13.5698  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    14.8887  -19.9323    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    16.2077  -19.1708    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    17.5266  -19.9323    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    10.9319  -17.6478    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0
     9.6129  -19.9323    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0
    12.2508  -21.4553    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0
   1  2  1  0        0
   2  3  1  0        0
   3  4  1  0        0
   4  5  1  0        0
   5  6  1  0        0
   1  7  2  0        0
   1  8  1  0        0
   2  9  1  0        0
M  END
 >  <Name> (1)
(+/-)2-BROMOHEXANOIC ACID  (2-BROMOCAPROIC ACID)

 >  <Amount> (1)
25G

 >  <Supplier catalogue number> (1)
24,283-7

 >  <Location> (1)
STORE

 >  <CAS_Number> (1)
2681-83-6

 >  <MW> (1)
195.0449

$$$$


thanks,
Martin

Dr. Martin Stoermer
Institute For Molecular Bioscience
University of Queensland

Hi Dave,

As intuitively as could be imagined:

bond n. fe, r. hem and n. n

provided that these selections match your atoms. You can also pick
both atoms with the mouse and then simply issue 'bond'.

Cheers,

Tsjerk

On Sat, May 24, 2008 at 7:50 AM, Charles David Stout <da...@sc...> wrote:
> How does one bond two atoms, such as Fe and heme N, without having all the
> CONECT records in the PDB file?
> Thanks
> Dave Stout
> 5-23-08
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>
>



-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623

How does one bond two atoms, such as Fe and heme N, without having all 
the CONECT records in the PDB file?
Thanks
Dave Stout
5-23-08

Hi Gary,

On Tue, 20 May 2008 12:12:43 +0200, Gary Hunter
<gar...@um...> wrote:

> I am displaying pdb files of metalloenzymes, and like to show the metal ions
> as spheres.
> I select the metal and use show>spheres
> Fine
> Somehow recently I must have done something to reset the vdw radius of
> manganese (MN in the pdb), as Mn atoms now display as small spheres (a
> little larger than solvent as nb_spheres).
> They were fine originally (!) and other metal atoms display at correct sizes
> (much bigger than my Mn).
> What could I have done?
> 
> A global edit of sphere_scale etc. simply changes all.
> I reset the setting>edit_all using the dialog as a check too.
> I can use:
> Alter name MN, vdw=vdw*2
> Rebuild
> 
> Or whatever, but I have to do it for each pdb file I read in now (if it has
> a Mn atom).
> Could I have messed with the vdw value in some way?
> 
> Is there a way to actually check the values being used for vdw radii?

You can get the vdw radii for any items (say you have a selection called
"metals") with:

  iterate metals, print chain, resn,resi,vdw

which will print the chain ID, residue name, residue number and van der Waals
radius for each.  Of course, this only applies to the vdw radius that is
assigned after you read in the PDB file.

Cheers,
Rob
-- 
Robert L. Campbell, Ph.D.
Senior Research Associate/Adjunct Assistant Professor 
Botterell Hall Rm 644
Department of Biochemistry, Queen's University, 
Kingston, ON K7L 3N6  Canada
Tel: 613-533-6821            Fax: 613-533-2497
<rob...@qu...>    http://pldserver1.biochem.queensu.ca/~rlc

Gary Hunter wrote:
> I am displaying pdb files of metalloenzymes, and like to show the 
> metal ions as spheres.
> I select the metal and use show>spheres
> Fine
> Somehow recently I must have done something to reset the vdw radius of 
> manganese (MN in the pdb), as Mn atoms now display as small spheres (a 
> little larger than solvent as nb_spheres).
> They were fine originally (!) and other metal atoms display at correct 
> sizes (much bigger than my Mn).
> What could I have done?
>
> A global edit of sphere_scale etc. simply changes all.
> I reset the setting>edit_all using the dialog as a check too.
> I can use:
> Alter name MN, vdw=vdw*2
> Rebuild
>
> Or whatever, but I have to do it for each pdb file I read in now (if 
> it has a Mn atom).
> Could I have messed with the vdw value in some way?
>
> Is there a way to actually check the values being used for vdw radii?
>
>
>
>
> Thanks
> Gary
>
>
>
>
>
>
>
>
>
>
>  Prof. Gary J. Hunter,
>  Department of Physiology and Biochemistry,
>  University of Malta, Msida, MSD 06, Malta.
>  phone: +356 2340 2917
>  phone: +356 21316655 (secretary),  Fax: +356 21310577
Gary,

You will normally have to manually adjust the ionic radius of metal ions 
to correctly render them. For starters, PDB files are not always good at 
distinguishing between atoms and ions; second, there is no single "ionic 
radius" for metal ions, as their effective radius depends on 
coordination number. (For example, zinc has different ionic radii for 4- 
and 5-coordinate compounds. A simple way to adjust radii globally by 
element is:

alter elem zn, vdw=0.74
rebuild

(This would set  zinc in a protein structure to 0.74A, the appropriate 
value for a four-coordinate metal environment.)

There are many databases out there for obtaining ionic radii. For 
example, see http://abulafia.mt.ic.ac.uk/shannon/ .

Cheers,


-- 
------------------------------------------------------------------------
Roger S. Rowlett
Professor
Colgate University Presidential Scholar
Department of Chemistry
Colgate University
13 Oak Drive
Hamilton, NY 13346

tel: (315)-228-7245
ofc: (315)-228-7395
fax: (315)-228-7935
email: rro...@ma...

Gary,

Sorry to hear you couldn't get CE to work for you.  If you have time, please 
email me (personally or via the PyMOLWiki) and let me know what frustrations 
you had.  I'd like to make it more user friendly.

To quickly answer your question about visualizing the true-CE results use 
split_states:

  # (1) load the CE result file
  load /path/to/ceResults.txt

  # (2) call split_states to show the other molecule
  split_states ceResults

HTH,

-- Jason

-- 

Jason Vertrees, PhD (javertre@{utmb,cs.dartmouth}.edu)

http://www.best.utmb.edu/
http://www.pymolwiki.org/

I am displaying pdb files of metalloenzymes, and like to show the metal ions
as spheres.
I select the metal and use show>spheres
Fine
Somehow recently I must have done something to reset the vdw radius of
manganese (MN in the pdb), as Mn atoms now display as small spheres (a
little larger than solvent as nb_spheres).
They were fine originally (!) and other metal atoms display at correct sizes
(much bigger than my Mn).
What could I have done?

A global edit of sphere_scale etc. simply changes all.
I reset the setting>edit_all using the dialog as a check too.
I can use:
Alter name MN, vdw=vdw*2
Rebuild

Or whatever, but I have to do it for each pdb file I read in now (if it has
a Mn atom).
Could I have messed with the vdw value in some way?

Is there a way to actually check the values being used for vdw radii?




Thanks
Gary










 Prof. Gary J. Hunter,
 Department of Physiology and Biochemistry,
 University of Malta, Msida, MSD 06, Malta.
 phone: +356 2340 2917
 phone: +356 21316655 (secretary),  Fax: +356 21310577

> Is there some way to increase the "perspective-ness" of the molecule 
> being displayed (more so than the default perspective mode),

I don't know whether this will work with the tracking software, but:

set field_of_view, 50
zoom

may do this trick.

Cheers,
Warren
--
DeLano Scientific LLC
Subscriber Support Services
mailto:su...@de...

-----Original Message-----
From: pym...@li...
[mailto:pym...@li...] On Behalf Of Sabuj
Pattanayek
Sent: Wednesday, May 14, 2008 2:50 PM
To: pym...@li...
Subject: Re: [PyMOL] [ccp4bb] While on the subject of stereo

Hi,

Warren DeLano wrote:
> This has already been done with PyMOL.  There's a video at:
>  
> http://molviz.cs.toronto.edu/molviz/
>  
> and the code is downloadable.
>  
> The stereo effect isn't so great with both eyes open, but I do think there
is potential for use of head or object tracking as a means of controlling
rotation.

I'm not sure what the technical term for this effect is, but is there some
way to increase the "perspective-ness" of the molecule being displayed (more
so than the default perspective mode), i.e. atoms in the back are "smaller"
than they should be and atoms near the front are "larger" than they should
be in vs an orthographic projection? This might increase the stereo effect
when moving from side to side.

Anyone found optimal settings in the molviz control panel so as to reduce
jitter?

Thanks,
Sabuj

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges.
Microsoft(R) Visual Studio 2008. 
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
PyMOL-users mailing list
PyM...@li...
https://lists.sourceforge.net/lists/listinfo/pymol-users

Hi,

Warren DeLano wrote:
> This has already been done with PyMOL.  There's a video at:
>  
> http://molviz.cs.toronto.edu/molviz/
>  
> and the code is downloadable.
>  
> The stereo effect isn't so great with both eyes open, but I do think there is potential for use of head or object tracking as a means of controlling rotation.

I'm not sure what the technical term for this effect is, but is there 
some way to increase the "perspective-ness" of the molecule being 
displayed (more so than the default perspective mode), i.e. atoms in the 
back are "smaller" than they should be and atoms near the front are 
"larger" than they should be in vs an orthographic projection? This 
might increase the stereo effect when moving from side to side.

Anyone found optimal settings in the molviz control panel so as to 
reduce jitter?

Thanks,
Sabuj

Dear all
I made a structural alignment using CE on-line resources (I cant get the
plugin etc to work in pymol !)
The output is a pdb file with two model structures (also designated chains A
and B respectively).
Pymol reads in the sequences fine, I can display the chains and select them,
but I can only visualise the first model in the structure display window.
When I select the first chain I see the slection on the structure in the
display window. When I select the second chain only the sequence display
shows that anything is selected.
I¹ve tried selecting it and Œshow lines¹ etc etc to no effect.
Am I missing a setting or something?
Gary








 Prof. Gary J. Hunter,
 Department of Physiology and Biochemistry,
 University of Malta, Msida, MSD 06, Malta.
 phone: +356 2340 2917
 phone: +356 21316655 (secretary),  Fax: +356 21310577