pymol-users Mailing List for PyMOL Molecular Graphics System

Matt,

 

Please note that we have not taken anything away!  MacPyMOL simply is
not (and has never been a Tcl/Tk/Tkinter)-based application in its
default configuration.  

 

To run it as an X11-based Tcl/Tk:  First, make sure that X11 is
installed and launched.  Then make a copy of MacPyMOL and rename it:

 

PyMOLX11Hybrid

 

When double-clicked, this renamed build will then launch using the same
Tcl/Tk/tkinter-based interface as the Windows and Linux versions.

 

Cheers,

Warren

 

________________________________

From: Matthew O'Meara [mailto:mo...@em...] 
Sent: Friday, January 30, 2009 6:45 AM
To: pym...@li...
Cc: Jon Latane
Subject: [PyMOL] plugins for MacPyMOL?

 

Hi,

 

We have created a wizard that we would like to distribute, however a
fair number of our users will be using the MacPyMOL version.  It seems
that the Plugin menu has been disabled as well as running scripts from
the "file->run..." menu.  Also plugin scripts dropped by hand into the
modules/pmg_tk/startup folder are not executed.  Why has it been
disabled?  Is there anyway that I can allow users to install our wizard?


 

Thanks,

 

Matt      

Hi,
We have created a wizard that we would like to distribute, however a fair
number of our users will be using the MacPyMOL version.  It seems that the
Plugin menu has been disabled as well as running scripts from the
"file->run..." menu.  Also plugin scripts dropped by hand into the
modules/pmg_tk/startup folder are not executed.  Why has it been disabled?
 Is there anyway that I can allow users to install our wizard?

Thanks,

Matt

Dear all,

Once again, I need some help from pymol experts.  I would like to select
surface residues via the surface calculation algorithm of Pymol.  Is this
possible?  What is the surface calculation algorithm that Pymol use and is
it scattered among many source-code files or is it contained in just one
source-code file?

Thanks in advance.
Best,
Quyen

On 13:10 Thu 29 Jan     , Abhinav Kumar wrote:
> Is there someway to select and show conserved residues when a bunch of 
> superimposed structures are loaded into Pymol?
> And if so, can identical and similar residues be selected separately?

I often use ConSurf, which will give you a PDB with a conservation index 
in the B-factor field. You can then load that into PyMol and color by B 
as usual.

Since it's unclear exactly what you intend to do with this, I can't tell 
you whether ConSurf makes sense for your use.

-- 
Thanks,
Donnie

Donnie Berkholz
Developer, Gentoo Linux
Blog: http://dberkholz.wordpress.com

Hi Abhinav,

On Thu, 29 Jan 2009 13:10:25 -0800, Abhinav Kumar
<abh...@sl...> wrote:

> Is there someway to select and show conserved residues when a bunch of 
> superimposed structures are loaded into Pymol?
> And if so, can identical and similar residues be selected separately?

I don't believe that there is a direct way to do it from the
superimposition, but I commonly apply "external data" to the B-factor
column of a PDB file for colouring purposes. For example, you could
calculate the degree of sequence conservation at each residue and colour
on that. I use the following scripts:

http://pldserver1.biochem.queensu.ca/~rlc/work/scripts/variability.py
http://pldserver1.biochem.queensu.ca/~rlc/work/scripts/seq_convert.py
http://pldserver1.biochem.queensu.ca/~rlc/work/pymol/data2bfactor.py

See the pages:

http://pldserver1.biochem.queensu.ca/~rlc/work/scripts/
  and
http://pldserver1.biochem.queensu.ca/~rlc/work/pymol

for more information.

When you align the structures, if you use the "object=" keyword (or do
it using the "align" menu) it will create an alignment object that you
can save as a Clustal format file. This file can then be used to calculate the
sequence variability. 

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

Dear all,

I was wondering what is the state of the art for this old dark art ...  
are there any good servers / programs that allow to easily upload your  
own sequence alignments or create a 'transparent' alignment (I want to  
see the alignment first and not a total black box) and then allow you  
to write out sequence conservation based either on identity or in e.g  
a Dayhoff matrix on the B factor column for displaying it later in eg  
Pymol?

To be clear I do not want a structural alignment, but mapping sequence  
alignment of eg a family to a single structure of a family member.

Thanks in advance, Tassos

Hi,

Is there someway to select and show conserved residues when a bunch of 
superimposed structures are loaded into Pymol?
And if so, can identical and similar residues be selected separately?

-- 
Thanks 
Abhinav 

Stanford Synchrotron Radiation Laboratory 
Joint Center for Structural Genomics 
Mail Stop 99 
Phone: (650) 926-2992 
Fax: (650) 926-3292 

Hi all,
I just upgraded to the pymol 1.2b2. The install works perfectly etc.
I have a strange phenomenon though, all the colors in the menu on the 
right side of the graphics windows have disappeared as well as the 
colors in the sequence panel.
Any hint what it might be?
I have not change my configuration (environment variables):
System: Windows XP sp3
Path containing C:\Program Files\Delano Scientific\Pymol
PYMOLPATH defined to C:\Program Files\Delano Scientific\Pymol
PYTHONPATH defined to C:\Python25 (I use python for other things...).

Other than that, everything works perfectly (as far as I have tested)

Gilles

Hi John,

There's a limit to what PyMOL can do by itself, but it sounds like
composition might be the easiest approach.  First, render the image
without the dashed line:

ray

save background.png

Then, disable everything but the dashed line and then render it alone
with opaque_background disabled:

unset opaque_background

ray

save foreground.png

Then combine the two images externally using something like GIMP,
Photoshop, or even PowerPoint.

cheers,
Warren

> -----Original Message-----
> From: John Pak [mailto:joh...@ut...]
> Sent: Monday, January 26, 2009 4:32 PM
> To: pym...@li...
> Subject: [PyMOL] Getting a dashed line through a transparent surface
> 
> Hi everyone.  First time poster, and I just want to say that Pymol's a
> great program! :)
> 
> I'm trying to get a dashed line (made via the Distance Wizard) from an
> atom outside of a rendered transparent surface (20% transparency) to
> an atom inside a rendered transparent surface, without a change in
> colour.  Right now, only the part of the dashed line that is outside
> the surface is the colour I set it to, whereas the part of the dashed
> line that is inside the surface is the colour of the surface.
> 
> This is for a figure for a paper, where I'm using the dashed line to
> indicate a a hydrogen bond.  Somehow, I want the dashed line to behave
> as if there is no transparent surface to pass through.
> 
> Thanks for the time.
> 
> 
>
------------------------------------------------------------------------
--
> ----
> This SF.net email is sponsored by:
> SourcForge Community
> SourceForge wants to tell your story.
> http://p.sf.net/sfu/sf-spreadtheword
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 
> 
> 

Hi everyone.  First time poster, and I just want to say that Pymol's a  
great program! :)

I'm trying to get a dashed line (made via the Distance Wizard) from an  
atom outside of a rendered transparent surface (20% transparency) to  
an atom inside a rendered transparent surface, without a change in  
colour.  Right now, only the part of the dashed line that is outside  
the surface is the colour I set it to, whereas the part of the dashed  
line that is inside the surface is the colour of the surface.

This is for a figure for a paper, where I'm using the dashed line to  
indicate a a hydrogen bond.  Somehow, I want the dashed line to behave  
as if there is no transparent surface to pass through.

Thanks for the time.

Hi Warren et al.,

I have been using Pymol for quite a while (years) and have not seen documentation on these two questions.  This is probably because I am still using rc0.99, but I promise to switch to the $$ release as soon as my grant comes in!

1st question:  Is there a way to selectively clip?  For example, I am showing a surface rendering of the inside of a binding pocket and the ligand inside.  I want to clip away the front half of the pocket, leaving the ligand intact.  I loaded my structure containing the ligand and deleted the ligand atoms. I then loaded a copy of my structure (renamed) and deleted everything in the second loaded structure but the ligand.  I tried masking the ligand, but that apparently only masks selections. Is there a way to "mask" the ligand from clipping.  I can do this in BallView, but it does not load the C-C bonds of the ligand correctly.  Besides, I prefer working in Pymol.

2nd question:  I recently gave a PowerPoint presentation to our department and showed the structure using the *.psw feature and an embedded hyperlink in the presentation.  I then was able to toggle through F1, F2, F3 etc...  using PgDn.  It worked great and got an enthusiastic response. I guess this is a PowerPoint question, but is there a way to keep PowerPoint from displaying the warning message before opening the file?  That is the only time I needed to go back to the podium during my talk.

Thanks.

__________________________________
Eric Peterson, PhD
Department of Pharmacology and Toxicology
Center for Alcohol and Drug Abuse Studies
College of Medicine
University of Arkansas for Medical Sciences


Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information.  Any unauthorized review, use, disclosure or distribution is prohibited.  If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.

On 23:26 Thu 22 Jan     , Warren DeLano wrote:
> I was very much hoping that this new display would nevertheless also
> work with existing nVidia Quadro-based Mac or Linux systems with
> existing emitters and glasses running existing OpenGL software.  Sadly,
> this does not seem to be the case due at least in part to the fact that
> the phase of the sync signal coming out of the Quadro card does not
> match the update phase of the LCD display.  In addition, light from the
> display itself seems to corrupt the sync signal for StereoGraphics
> glasses.
> 
> So, in summary, it seems we are out of luck until nVidia does a bit more
> work.  Reportedly, there is a driver update coming that will target use
> of nVidia's 3D vision glasses with Quadro-based graphics cards, so stay
> tuned.  

Just in case anyone wants to track this, here is the source I found of 
that report:

"Quad Buffered Stereo will be supported with the NVIDIA GeForce 3D 
vision glasses in upcoming driver release by spring 2009."

<" rel="nofollow">http://www.nvidia.com/object/quadro_stereo_technology.html>

-- 
Thanks,
Donnie

Donnie Berkholz
Developer, Gentoo Linux
Blog: http://dberkholz.wordpress.com

Stereo 3D Users:

I tested out the Samsung Syncmaster 2233RZ / NVIDIA 3D Vision bundle
today ($599 for a 120 Hz LCD display with one pair of glasses).

Unfortunately, the 2233RZ display is clearly not yet a drop-in
replacement for your stereo-3D-capable CRT:  The display is very nice,
and the stereo quality is excellent (as good as a CRT!), but only
provided that you:

1. Run Microsoft Vista.

2. Install a high-end GeForce 3D card (9800 GT or similar).

3. Have DirectX-based software than has been modified to support
nVidia's "consumer-grade" stereo API.

4. Use nVidia's 3D Vision glasses and sync emitter (a USB-based device).

In other words, despite the high stereo image quality, this is not yet a
servicable stereo 3D solution for professional use.  Rats!

I was very much hoping that this new display would nevertheless also
work with existing nVidia Quadro-based Mac or Linux systems with
existing emitters and glasses running existing OpenGL software.  Sadly,
this does not seem to be the case due at least in part to the fact that
the phase of the sync signal coming out of the Quadro card does not
match the update phase of the LCD display.  In addition, light from the
display itself seems to corrupt the sync signal for StereoGraphics
glasses.

So, in summary, it seems we are out of luck until nVidia does a bit more
work.  Reportedly, there is a driver update coming that will target use
of nVidia's 3D vision glasses with Quadro-based graphics cards, so stay
tuned.  

Cheers,
Warren

Hi Warren,

> I am pleased to learn that MacPyMOL does in fact work over NFS!  We're
> obviously subject to reporter bias in terms of hearing mostly about
> problems, not working solutions, but this certainly didn't work for us
> previously (on Panther perhaps?  Well before Leopard...)

I went through every old PyMOL ticket we had in our database (48, to be 
exact), and I found out about the issues with older PyMOL versions and NFS.
We integrate all the programs we support into a giant tcsh shell
initialization file, and prior to August 2007, we had an alias like this;

alias pymol 'open /path/to/pymol/directory'

which made no sense to me at the time, so I changed it to:

alias pymol 'open /path/to/pymol/MacPyMOL.app'

It turns out that we had been advising people to run the 'pymol' alias,
which would open the directory in Finder, then copy the .app to their 
hard drive to run it in order to workaround the aforementioned NFS issues.

I made the change to call the .app directly without knowing the history
there, and we have not gotten a single ticket on it regarding NFS issues
since then.  At the time, we were running an open source 0.99 build
compiled by David Gohara.  We have been using the incentive builds since we
signed up for them.

Anyway I feel pretty comfortable saying that MacPyMOL is working fine when 
run from an NFS volume these days.  AFP may be a different matter though...
:-)

> Rereading your emails more carefully, I think I understand now what
> you're angling at:  remote command-line use of MacPyMOL.  That shouldn't
> work...
> 
> Instead, you need a pure X11-based build of PyMOL.

Okay, that makes sense.  Actually, the ticket archives held a clue about 
this one as well.  I believe that my user was using a pure-X11 build 
circa 0.97 that was also built by David.  That ticket was from February
2005, so this was almost 4 years ago now.

I'll install the 1.2 betas and try them out.

Thanks for your help.

-ben

--
Ben Eisenbraun
Structural Biology Grid                           Harvard Medical School
http://sbgrid.org                                 http://hms.harvard.edu

Ben,

I am pleased to learn that MacPyMOL does in fact work over NFS!  We're
obviously subject to reporter bias in terms of hearing mostly about
problems, not working solutions, but this certainly didn't work for us
previously (on Panther perhaps?  Well before Leopard...)

Rereading your emails more carefully, I think I understand now what
you're angling at:  remote command-line use of MacPyMOL.  That shouldn't
work...

Instead, you need a pure X11-based build of PyMOL.  We haven't provided
those in the past (aside from ipymol), but I recently posted a couple of
builds for 1.2 beta 0:

http://delsci.com/ip/12/beta/

pymol-1_2b0-bin-osx-x86-x11-tiger.tgz
 
 - or -

pymol-1_2b0-bin-osx-ppc-x11-tiger.tgz

There are "pure" linux-like PyMOL builds for Mac OS X / X11.

Cheers,
Warren

> -----Original Message-----
> From: Ben Eisenbraun [mailto:be...@cr...]
> Sent: Thursday, January 22, 2009 9:14 AM
> To: Warren DeLano
> Cc: PyM...@li...
> Subject: Re: [PyMOL] Command line mode on OS X 10.4 vs 10.5
> 
> Hi Warren,
> 
> > MacPyMOL has always had problems running off of a remote NFS drive
--
> > we've never been able to figure out why, but it is clearly something
to
> > do with the magic behind Apple's application bundles, since ordinary
X11
> > builds run just fine.
> 
> Interesting.  We've been using it installed on an NFS volume for
several
> years now, and searching our ticket database didn't find any PyMOL
issues
> that looked related to the network volume.  That said, I have
definitely
> seen oddities on Macs with our set up, although I suspect some of that
is
> caused by the case-sensitivity issue more than anything else.
> 
> > The workaround is to copy the MacPyMOL bundle onto a local drive
(/tmp?)
> > before running it.  Can you have your user try that?
> 
> I tried this, but got the same error.
> 
>  $ /tmp/test/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
> _RegisterApplication(), FAILED TO establish the default connection to
the
> WindowServer, _CGSDefaultConnection() is NULL.
> 
> -ben
> 
> --
> Ben Eisenbraun
> Structural Biology Grid                           Harvard Medical
School
> http://sbgrid.org
http://hms.harvard.edu
> 
> 
> 

Hi Warren,

> MacPyMOL has always had problems running off of a remote NFS drive --
> we've never been able to figure out why, but it is clearly something to
> do with the magic behind Apple's application bundles, since ordinary X11
> builds run just fine.

Interesting.  We've been using it installed on an NFS volume for several
years now, and searching our ticket database didn't find any PyMOL issues 
that looked related to the network volume.  That said, I have definitely
seen oddities on Macs with our set up, although I suspect some of that is 
caused by the case-sensitivity issue more than anything else.

> The workaround is to copy the MacPyMOL bundle onto a local drive (/tmp?)
> before running it.  Can you have your user try that?

I tried this, but got the same error.

 $ /tmp/test/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
_RegisterApplication(), FAILED TO establish the default connection to the
WindowServer, _CGSDefaultConnection() is NULL.

-ben

--
Ben Eisenbraun
Structural Biology Grid                           Harvard Medical School
http://sbgrid.org                                 http://hms.harvard.edu

Ben,

MacPyMOL has always had problems running off of a remote NFS drive -- we've never been able to figure out why, but it is clearly something to do with the magic behind Apple's application bundles, since ordinary X11 builds run just fine.

The workaround is to copy the MacPyMOL bundle onto a local drive (/tmp?) before running it.  Can you have your user try that?

Cheers,
Warren


-----Original Message-----
From: Ben Eisenbraun [mailto:be...@cr...]
Sent: Thu 1/22/2009 7:31 AM
To: Warren DeLano
Cc: PyM...@li...
Subject: Re: [PyMOL] Command line mode on OS X 10.4 vs 10.5
 
Hi Warren,

> /Applications/MacPyMOL.app/Contents/MacOS/MacPyMOL -qc 
> 
> seems to work on our Leopard machines.  Is /programs on a shared drive
> by any chance?

Yes, it is.  It's an NFS volume hosted on a linux server, so in addition to 
being a network volume, it's also case-sensitive.

Some additional details: the use case is that my user has an Xserve that he
ssh's into and runs jobs on without being logged on to the console.  He
recalls being able to do this with an older version of MacPyMOL and/or OS
X.

>From what I can see testing here, this doesn't work at all if you're not 
logged into the console.  Is it supposed to?

Thanks.

-ben

--
Ben Eisenbraun
Structural Biology Grid                           Harvard Medical School
http://sbgrid.org                                 http://hms.harvard.edu

Vince,

I started to compile and answer for you, but then it got too long to be
appropriate for the list.  For a discussion of "fit" and optimal
alignment please see the wiki page:

  http://www.pymolwiki.org/index.php/Cealign#Fit_vs._optAlign

Briefly, PyMOL's "fit" is fast and works well.  I believe it's based off
the Jacobian, which iteratively arrives at the solution (to within some
tiny  error).  If you have to use something with a known reference then
check out the "optAlign" function from the qkabsch.py file that comes
with the CEalign package:

  http://www.pymolwiki.org/index.php/Cealign

If not, you can just use "fit" and avoid installing new software.  :-)

HTH,

-- Jason


> Message: 4
> Date: Tue, 20 Jan 2009 12:29:20 -0800
> From: "Vincent Chu" <vin...@st...>
> Subject: [PyMOL] Algorithm for pyMOL's fit and rmsd algorithms
> To: pym...@li...
> Message-ID:
>         <441...@ma...>
> Content-Type: text/plain; charset=ISO-8859-1
> 
> Hi all --
> 
> I've been using pyMOL's fit and rmsd commands on small numbers of
> structures but now I'd like to use these functions on large numbers of
> molecules.
> 
> Does anybody know what algorithm is used in pyMOL to superimpose
> molecular structures on top of other ones (i.e., pymol's fit
> function). Better yet would be a reference.
> 
> Thanks!
> 
> Vince


-- 

Jason Vertrees, PhD

Dartmouth College : jv...@cs...
Boston University : ja...@bu...

PyMOLWiki : http://www.pymolwiki.org/

Hi Warren,

> /Applications/MacPyMOL.app/Contents/MacOS/MacPyMOL -qc 
> 
> seems to work on our Leopard machines.  Is /programs on a shared drive
> by any chance?

Yes, it is.  It's an NFS volume hosted on a linux server, so in addition to 
being a network volume, it's also case-sensitive.

Some additional details: the use case is that my user has an Xserve that he
ssh's into and runs jobs on without being logged on to the console.  He
recalls being able to do this with an older version of MacPyMOL and/or OS
X.

>From what I can see testing here, this doesn't work at all if you're not 
logged into the console.  Is it supposed to?

Thanks.

-ben

--
Ben Eisenbraun
Structural Biology Grid                           Harvard Medical School
http://sbgrid.org                                 http://hms.harvard.edu

Ben,

Hmm...

/Applications/MacPyMOL.app/Contents/MacOS/MacPyMOL -qc 

seems to work on our Leopard machines.  Is /programs on a shared drive
by any chance?

Cheers,
Warren

> -----Original Message-----
> From: Ben Eisenbraun [mailto:be...@cr...]
> Sent: Wednesday, January 21, 2009 10:38 AM
> To: PyM...@li...
> Subject: [PyMOL] Command line mode on OS X 10.4 vs 10.5
> 
> 
> Howdy,
> 
> Under OS X 10.4, I can do this:
> 
> [bene@flange][/nfs/home/bene] /programs/i386-
> mac/pymol/1.1r1/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
> [bene@flange][/nfs/home/bene] echo $?
> 0
> 
> But on 10.5, I get this:
> 
> [bene@macintel27][/nfs/home/bene] /programs/i386-
> mac/pymol/1.1r1/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
> _RegisterApplication(), FAILED TO establish the default connection to
the
> WindowServer, _CGSDefaultConnection() is NULL.
> 2009-01-21 13:31:53.279 MacPyMOL[38070:807] An uncaught exception was
> raised
> 2009-01-21 13:31:53.280 MacPyMOL[38070:807] Error (1002) creating
> CGSWindow
> 2009-01-21 13:31:53.280 MacPyMOL[38070:807] *** Terminating app due to
> uncaught exception 'NSInternalInconsistencyException', reason: 'Error
> (1002) creating CGSWindow'
> 
> Is this a known issue?  Is there a different switch that has to be
passed
> for 10.5?
> 
> Thanks.
> 
> -ben
> 
> --
> Ben Eisenbraun
> Structural Biology Grid                           Harvard Medical
School
> http://sbgrid.org
http://hms.harvard.edu
> 
>
------------------------------------------------------------------------
--
> ----
> This SF.net email is sponsored by:
> SourcForge Community
> SourceForge wants to tell your story.
> http://p.sf.net/sfu/sf-spreadtheword
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 
> 

Hi David,

# With regard to question 1, issue:

window position,0,0

# followed by

window size,7860,2160

# With regard to question 2, launch with the -j option to get the gui
drawn twice, or issue:

stereo geowall

# and if necessary:

stereo swap

# With regard to question 3, we don't have a solution yet, but we are
going to have to solve this someday soon!

Cheers,
Warren



> -----Original Message-----
> From: David Borland [mailto:bo...@re...]
> Sent: Tuesday, January 20, 2009 1:14 PM
> To: pym...@li...
> Subject: [PyMOL] PyMOL in stereo at UNC Chapel Hill
> 
> Hello.  I am a Senior Visualization Researcher with The Renaissance
> Computing Insitute at UNC-Chapel Hill (www.renci.org).  I have
recently
> begun working with a couple of groups at UNC, including Brian
Kuhlman's
> in conjunction with the Rosetta protein design and prediction project,
> to enable them to run PyMOL on our Ultra-high resolution stereo
> display.  Our display consists of 2 Sony 4k (4xHD resolution)
projectors
> set up for rear projection.  The projectors fully overlap, and use
> circularly polarizing filters to achieve passive stereo.  We are
running
> on Windows XP.
> 
> So far, without touching any code, I can get it running in our
> environment by dragging the window to the top-left corner of the
display
> (would prefer a borderless window positioned at (0,0), but moving the
> left edge just off the screen works), and calling viewport 7860,2160
> (the total resolution of the two projectors, side-by-side).  I can
then
> turn on wall-eye stereo, load a file, and view it in stereo.
> 
> There are a number of issues with this approach, however:
> 
> 1.  I need to manually move the window just off the left edge of the
> screen.  I'm aware that I can set the window position as a command
line
> argument, but this will set the window border to the top-left corner,
> which is not quite what I want.  Glut will not let us get a borderless
> window (it supports fullscreen mode, but this will only fullscreen to
> the primary monitor on windows, and the 2 projectors are seen as 8
> monitors by Windows).  However there is a hack to get around this.
> Calling glutPositionWindow(0, 0) after creating the window will move
the
> client area of the window to (0, 0), which is what we want (I have
> tested this with p_glutPositionWindow(0, 0) from within PyMOL).
> 
> 2.  The internal gui is problematic.  We basically need need both
> projectors to project the same image, except for the difference in
> viewpoint for the stereo model.  Pushing the internal gui off the side
> of the display gives correct stereo rendering, but makes the gui
> unusable.  Bringing the internal gui back onto the screen by changing
> the view width moves the stereo images out of alignment.  The ideal
> solution here would be to fully replicate the internal gui in both the
> left and right halves of the screen.  This way, the two stereo images
> would line up correctly, and both the left and right eyes would see a
> consistent view of the internal gui.  It would be nice if both copies
of
> the interal gui could be manipulated, as then it wouldn't matter which
> half of the screen your mouse happened to be on.  A similar thing
> (replicate for left and right views) could also be done for the
command
> line at the bottom of the screen.
> 
> 3.  A lesser problem is the fact that due to the ultra-high resolution
> of our displays, the text (both in the guis and in the 3D view) can be
> hard to read.  It would be nice to be able to adjust this.
> 
> 
> 
> I was hoping to get some feedbackregarding what you think might be the
> best approach to tackle these issues (especially #2) and integrate
these
> changes into PyMOL.  I think such a stereo mode could be a nice
> contribution to PyMOL, as this type of dual-display stereo is quite
> common, and I'd be happy to make any changes available to the PyMOL
> community.
> 
> I hope this all makes sense, and thanks in advance for any feedback.
> 
> 
> David Borland
> 
> 
> 
> 
> 
> 
>
------------------------------------------------------------------------
--
> ----
> This SF.net email is sponsored by:
> SourcForge Community
> SourceForge wants to tell your story.
> http://p.sf.net/sfu/sf-spreadtheword
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 
> 

Vince,

I apologize, but I do not know its ultimate source -- that minimal RMSD
fitting code has been floating around my toolkit for years.  My personal
belief is that the approach is equivalent to Kabsch yet stable with
respect to degenerate cases (e.g. all points in a plane).  However, I
have neither the linear algebra chops nor the time to prove it.

The reference for Kabsch is here:

Kabsch, Wolfgang, (1976) "A solution of the best rotation to relate two
sets of vectors", Acta Crystallographica 32:922.

Of course, if anyone ever comes across a case where PyMOL performs
noticably worse than Kabsch, we would certainly like to know about it!  

Cheers,
Warren

> -----Original Message-----
> From: Vincent Chu [mailto:vin...@st...]
> Sent: Tuesday, January 20, 2009 1:14 PM
> To: pym...@li...
> Subject: [PyMOL] Algorithm for pyMOL's fit and rmsd algorithms
> 
> Hi all --
> 
> I've been using pyMOL's fit and rmsd commands on small numbers of
> structures but now I'd like to use these functions on large numbers of
> molecules.
> 
> Does anybody know what algorithm is used in pyMOL to superimpose
> molecular structures on top of other ones (i.e., pymol's fit
> function). Better yet would be a reference.
> 
> Thanks!
> 
> Vince
> 
> --
> ------------------------
> Vincent Chu
> Department of Applied Physics
> Geballe Laboratory of Advanced Materials
> McCullough Bldg. 318
> 476 Lomita Mall
> Stanford, CA, 94305-4045
> 
> vincentc [at] stanford.edu
> http://www.stanford.edu/~vincentc/
> 415.279.7733
> 
> "Join a group. Get an identity."
> 
>
------------------------------------------------------------------------
--
> ----
> This SF.net email is sponsored by:
> SourcForge Community
> SourceForge wants to tell your story.
> http://p.sf.net/sfu/sf-spreadtheword
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 
> 

Sébastien,

 

PyMOL doesn't do this, in part because atoms in any given PDB file output from PyMOL could in principle derive from any loadead structure, from more than one structure, or be completely made up, and we wouldn't want incorrect or conflicting meta information out there polluting the net.  Nevertheless, at present, PyMOL cannot even support the use case where all of the output atoms are indeed derived from a single PDB file with associated headers.  Thus, you will need to transfer them manually.

 

Cheers,

Warren

 

________________________________

From: Sébastien Conilleau [mailto:sco...@ch...] 
Sent: Tuesday, January 20, 2009 7:56 AM
To: pym...@li...
Subject: [PyMOL] writing pdb files with headers

 

Hi,

I noticed that pymol omits the headers (I mean the 5 first line) when you try to write a pdb file. At this point I do it using this command line in my script :

> cmd.save(newName+".pdb","tmp")

Is there an implemeted way to avoid this behaviour or do I have to play around with files ?

Thanks

-- 

Sébastien Conilleau, PhD

UMR CNRS 7177

Laboratoire d'Infochimie

Université Louis Pasteur

4, rue Blaise Pascal

67008 Strasbourg Cedex, France

Tel : +(33)3-90-24-15-63

 

 

Howdy,

Under OS X 10.4, I can do this:

[bene@flange][/nfs/home/bene] /programs/i386-mac/pymol/1.1r1/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
[bene@flange][/nfs/home/bene] echo $?
0

But on 10.5, I get this:

[bene@macintel27][/nfs/home/bene] /programs/i386-mac/pymol/1.1r1/PyMOLX11Hybrid.app/Contents/MacOS/MacPyMOL -qc
_RegisterApplication(), FAILED TO establish the default connection to the WindowServer, _CGSDefaultConnection() is NULL.
2009-01-21 13:31:53.279 MacPyMOL[38070:807] An uncaught exception was raised
2009-01-21 13:31:53.280 MacPyMOL[38070:807] Error (1002) creating CGSWindow
2009-01-21 13:31:53.280 MacPyMOL[38070:807] *** Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Error (1002) creating CGSWindow'

Is this a known issue?  Is there a different switch that has to be passed
for 10.5?

Thanks.

-ben

--
Ben Eisenbraun
Structural Biology Grid                           Harvard Medical School
http://sbgrid.org                                 http://hms.harvard.edu

Scott,

the MSEs are probably HETATM records instead of ATOM records. You can either replace the HETATM records to ATOM records or set surface_mode to 1 which allows rendering of surfaces with HETATM records (I think, am a bit fuzzy on that)

HTH

	Carsten

-----Original Message-----
From: Scott Lefurgy [mailto:sle...@ae...]
Sent: Wednesday, January 21, 2009 12:20 PM
To: pym...@li...
Subject: [PyMOL] selenomethionine



I am looking at a structure that contains selenomethionine (rendered MSE
in the sequence).  When I Show Surface, Se-Met is not displayed with a
surface, but all the other residues are.  Is there any way to make a
surface over Se-Met as well?

Scott

---------------------------------------
Scott Lefurgy, Ph.D
Department of Microbiology & Immunology
Albert Einstein College of Medicine
Yeshiva University
111 Ullmann Building
1300 Morris Park Ave.
Bronx, NY 10461
718-430-2858
sle...@ae...
---------------------------------------


------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________
PyMOL-users mailing list
PyM...@li...
https://lists.sourceforge.net/lists/listinfo/pymol-users