pymol-users Mailing List for PyMOL Molecular Graphics System

John,

	You're not doing anything wrong -- there's just a little confusion
here about how settings work: they're global for an object or for the whole
system.  The current value is what is always used.  To get different sphere
sizes, use the alter command to change the vdw radii of the atoms for which
you wish to change the radii, followed by a rebuild command.

   alter selection, expression
   rebuild

For example:

   alter name ca, vdw=vdw*0.5
   alter resi 200, vdw=vdw*2.0
   rebuild

Cheers,
Warren

> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> John Berrisford
> Sent: Thursday, February 26, 2004 2:13 AM
> To: pym...@li...
> Subject: [PyMOL] sphere sizes
> 
> I am trying to create a picture containing a metal ion and 
> water molecules.  The problem that I am having is that I wish 
> to display them both as spheres, but of different sizes. 
> Pymol (version 0.93 and 0.95 beta 11 ) seems to only use the 
> last sphere size command i.e. "set sphere_scale, 0.1" in the 
> script and effectively ignore any previous sphere_scale command.
> 
> Am I doing something wrong and there is another, better way 
> of doing this, for example linking the size command to an object?
> 
> I have been so far unable to work out how to do this.
> 
> Thanks
> 
> John
> 
> --
> John Berrisford
> Department of Molecular Biology and Biotechnology University 
> of Sheffield
> 
> email: j.b...@sh...
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Works great!  Many thanks!

Doug

On Feb 26, 2004, at 12:56 AM, wg...@ch... wrote:

> Howdie, Citizens:
>
> Jack Howarth very kindly got a current version of X-windows pymol 
> working on Mac OS X via fink.  These files work for 10.3 and python 
> 2.3, and he also has some that work with python2.2.x
> .....
> Hopefully someone will get this into fink within the next few days.  
> Please feel free to nag the maintainers.
>
> Bill

I am trying to create a picture containing a metal ion and water molecules.  The
problem that I am having is that I wish to display them both as spheres, but of
different sizes. Pymol (version 0.93 and 0.95 beta 11 ) seems to only use the
last sphere size command i.e. "set sphere_scale, 0.1" in the script and
effectively ignore any previous sphere_scale command.

Am I doing something wrong and there is another, better way of doing this, for
example linking the size command to an object?

I have been so far unable to work out how to do this.

Thanks

John

-- 
John Berrisford
Department of Molecular Biology and Biotechnology
University of Sheffield

email: j.b...@sh...

Howdie, Citizens:

Jack Howarth very kindly got a current version of X-windows pymol 
working on Mac OS X via fink.  These files work for 10.3 and python 
2.3, and he also has some that work with python2.2.x

http://chemistry.ucsc.edu/~wgscott/xtal/xtalfink/pymol.info
http://chemistry.ucsc.edu/~wgscott/xtal/xtalfink/pymol.patch

To install this, you will need the latest version of pmw (version 1.2), 
which is not yet in fink.  Apparently the
previous version has a bug.  You can install it using this:

http://chemistry.ucsc.edu/~wgscott/xtal/xtalfink/pmw-py23.info

Copy these 3 files into /sw/fink/dists/local/main/finkinfo  (you might 
need to create it first).

and then issue

fink install pymol

If anything is missing, try

fink selfupdate-cvs
and then either temporarily enable the "unstable" branch or copy the 
needed files over to your local directory.

Hopefully someone will get this into fink within the next few days.  
Please feel free to nag the maintainers.

Bill


William G. Scott

Associate Professor
Department of Chemistry and Biochemistry
and The Center for the Molecular Biology of RNA
Sinsheimer Laboratories
University of California at Santa Cruz
Santa Cruz, California 95064
USA

url:          http://chemistry.ucsc.edu/~wgscott/

Mads,

	The only way to accomplish this that I can think of would be to
perform the normal expansion of all symmetry-related molecules within a
cutoff and then use a Python script to delete those objects which don't fit
a criteria.

	One such criteria might be the presence of a tell-tale atom or group
in the same plane as the original (trivial if your cell axes are orthogol).

PyMOL functions you'll need:

   cmd.get_names('objects')
   cmd.get_model(object-name)
   cmd.delete(object-name)

Cheers,
Warren

> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> Mads Gabrielsen
> Sent: Monday, February 23, 2004 5:37 AM
> To: pym...@li...
> Subject: [PyMOL] Symmetry
> 
> I am trying to create an image of the packing of my structure.
> The ASU is a monomer and the symmetry creates the  larger 
> quternary complex. I would like to show the packing of these 
> larger complexes, but it is to messy if I draw symmtry 
> related molecules in all three dimenisons. I would like to be 
> able to draw objects on one plane only. 
> Is there an easy way of doing this in PyMol or do I need to 
> delete all the objects I do not want by hand?
> 
> Cheers,
> 
> 
> --
> ----------------------------------------------------------
> Mads Gabrielsen, MSc
> Research Assistant / PhD Student
> 
> WNH-group                               Phone:(0044) 01382 345 746 
> MSIWTB, School of Life Sciences         Fax: (0044) 01382 345 764
> University of Dundee                    E-mail: 
> m.g...@du...
> Scotland
> 
> 
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Dieter,

	Yes, there is a practical limit which depends on (1) the amount of
RAM and (2) the operating system.  For reasons I don't completely
understand, PyMOL isn't able to get as much memory under Windows as under
Linux or OSX on equivalent system.

There are some things you can do to simply the scene, depending on which
representation you're using:

For spheres:

  set cull_spheres,0
  set sphere_quality,0

For sticks:

  set stick_quality,5

For cartoons:

  set cartoon_sampling, 3

When ray-tracing, lowering hash_max decreases memory usage at the cost of
performance (and vice-versa).
  
  set hash_max, 70

Also, the most recent beta versions usage less memory when raytracing...

http://delsci.com/beta
http://delsci.com/macpymol

I think 400k atoms is reasonable for 512 MB.  Which virus structure file are
you trying to render?

Cheers,
Warren

> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> dieter blaas
> Sent: Tuesday, February 24, 2004 10:52 PM
> To: pym...@li...
> Subject: [PyMOL] Are there limits for the number of 
> coordinates accepted?
> 
> When I tried to render a virus surface (about 400 000 
> coordinates) using a PC with 512MB RAM Pymol crashed 
> immediately. Is there a limit for the number of coordinates 
> accepted? Do I have to decrease the resolution by some means? 
> Thanx, Dieter
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

When I tried to render a virus surface (about 400 000 coordinates) using 
a PC with 512MB RAM Pymol crashed immediately. Is there a limit for 
the number of coordinates accepted? Do I have to decrease the 
resolution by some means? Thanx, Dieter

I am trying to create an image of the packing of my structure.
The ASU is a monomer and the symmetry creates the  larger quternary 
complex. I would like to show the packing of these larger complexes, but 
it is to messy if I draw symmtry related molecules in all three 
dimenisons. I would like to be able to draw objects on one plane only. 
Is there an easy way of doing this in PyMol or do I need to delete all 
the objects I do not want by hand?

Cheers,


-- 
----------------------------------------------------------
Mads Gabrielsen, MSc
Research Assistant / PhD Student

WNH-group                               Phone:(0044) 01382 345 746 
MSIWTB, School of Life Sciences         Fax: (0044) 01382 345 764
University of Dundee                    E-mail: m.g...@du...
Scotland

John,

It is simply a matter of restricting your atom selections to lie within that
single object:

dist my_line, e4_final//A/97/OE1, e4_final//M/200/MN
hide labels

See

http://pymol.sourceforge.net/newman/user/S0220commands.html#7

for more information on atom selections.

Cheers,
Warren


> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> John Berrisford
> Sent: Friday, February 20, 2004 3:27 AM
> To: Warren DeLano; pym...@li...
> Subject: RE: [PyMOL] drawing hydrogen bonds or dashes
> 
> Thanks that works fine apart from one thing: 
> I wish to display lines between 97AOE1 and MN in two 
> overlapped structures, with identical chain identification, 
> in two different colours. Using the below command draws two 
> lines in the same colour. Is there a way a specifying a line 
> between the two below atoms only within a specific pdb file 
> (defined as an object).
> 
> eg line between A97OE1 and MN only within object e4_final.
> 
> I hope this is clear
> 
> thanks
> 
> John
> 
> Quoting Warren DeLano <wa...@de...>:
> 
> > John,
> > 
> > Welcome to PyMOL!
> > 
> > > I wish to do the equivalent to the following molscript 
> command "line 
> > > position res-atom A97 OE1 to position res-atom M200 MN;"
> > 
> >    dist my_line, a/97/OE1, M/200/MN
> >    hide labels
> > 
> > Also note that PyMOL can draw hydrogen-bonds automatically 
> using its 
> > limited chemical intuition according to the global "h_bond_..." 
> > settings. (Sorry guys, that isn't documented -- I am thinking about 
> > creating a new command for it since "dist" isn't the right 
> place for this).
> > 
> >    dist hbnd, all, all, mode=2
> >    hide labels
> > 
> > (But I don't think that will help with MN.)
> > 
> > If you don't want dashes, then
> > 
> >    set dash_length, 3
> >    set dash_gap, 0
> > 
> > Seems to work for me...
> > 
> > Cheers,
> > Warren
> > 
> > 
> > > -----Original Message-----
> > > From: pym...@li...
> > > [mailto:pym...@li...] On 
> Behalf Of John 
> > > Berrisford
> > > Sent: Thursday, February 19, 2004 10:09 AM
> > > To: pym...@li...
> > > Subject: [PyMOL] drawing hydrogen bonds or dashes
> > > 
> > > I have searched the manual and forum and cannot find how 
> to draw a 
> > > hydrogen bond between two specific atoms.
> > > 
> > > I wish to do the equivalent to the following molscript 
> command "line 
> > > position res-atom A97 OE1 to position res-atom M200 MN;"
> > > 
> > > which draws a line between two atoms A97 OE1 and Mn200. So not a 
> > > hydrogen bond but a specific line.
> > > 
> > > Is this possible in pymol?
> > > 
> > > I have read posts on changing the thickness of a dash and 
> the number 
> > > of dashes in a line. But none on how to draw the line.
> > > 
> > > Can anyone help me, i'm a new user by the way!
> > > 
> > > thanks
> > > 
> > > John
> > > 
> > > 
> > > 
> > > -------------------------------------------------------
> > > SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> > > Build and deploy apps & Web services for Linux with a free DVD 
> > > software kit from IBM. Click Now!
> > > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> > > _______________________________________________
> > > PyMOL-users mailing list
> > > PyM...@li...
> > > https://lists.sourceforge.net/lists/listinfo/pymol-users
> > > 
> > 
> > 
> > 
> 
> 
> --
> John Berrisford
> Department of Molecular Biology and Biotechnology University 
> of Sheffield
> 
> email: j.b...@sh...
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Paul,

      (Where you wrote mesh, you meant surface right?)

	PyMOL's OpenGL renderer does a poor job of combining transparent
objects, but the ray-tracer should be capable of doing this so long as
multi-layer transparency is enabled.  This may or may not be acceptable
given all additional geometric elements you will end up with as a result.

See example image at:

http://www.pymol.org/ex/040220A.png


To achieve this, adjust the settings as follows:

set transparency, 0.5
set cartoon_transparency, 0.9
set transparency_mode, 1
set backface_cull, 0
set two_sided_lighting, 1
set ray_shadows, 0

ray

Of course, an alternative is to really use a see-through mesh, instead of a
surface...

http://www.pymol.org/ex/040220B.png

Default settings +

set ray_shadows, 0
set min_mesh_spacing, 0.4
set cartoon_transparency, 0.55

Cheers,
Warren

http://delsci.com/macpymol


> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of Paul Wan
> Sent: Friday, February 20, 2004 7:37 AM
> To: pym...@li...
> Subject: [PyMOL] Cartoon Transparency AND Transparent Mesh
> 
> Hi
> Just got my hands on Pymol G5 version and having loadsa fun with it.
> I'm trying to make a diagram and wondered if it is possible 
> to make the cartoon transparent then put a mesh over the 
> whole thing (also made transparent).
> 
> Currently I can make th ecartoon transparent but as soon as I 
> add a mesh (no matter what the transparency level) I loose 
> the Cartoon ribbon.
> Is there anyone out there that can help me do both?
> Thanks
> Paul
> 
>

Hi
Just got my hands on Pymol G5 version and having loadsa fun with it.
I'm trying to make a diagram and wondered if it is possible to make the
cartoon transparent then put a mesh over the whole thing (also made
transparent).

Currently I can make th ecartoon transparent but as soon as I add a mesh
(no matter what the transparency level) I loose the Cartoon ribbon.
Is there anyone out there that can help me do both?
Thanks
Paul

Thanks that works fine apart from one thing: 
I wish to display lines between 97AOE1 and MN in two overlapped structures, with
identical chain identification, in two different colours. Using the below
command draws two lines in the same colour. Is there a way a specifying a line
between the two below atoms only within a specific pdb file (defined as an
object).

eg line between A97OE1 and MN only within object e4_final.

I hope this is clear

thanks

John

Quoting Warren DeLano <wa...@de...>:

> John,
> 
> Welcome to PyMOL!
> 
> > I wish to do the equivalent to the following molscript 
> > command "line position res-atom A97 OE1 to position res-atom M200 MN;"
> 
>    dist my_line, a/97/OE1, M/200/MN
>    hide labels
> 
> Also note that PyMOL can draw hydrogen-bonds automatically using its limited
> chemical intuition according to the global "h_bond_..." settings. (Sorry
> guys, that isn't documented -- I am thinking about creating a new command
> for it since "dist" isn't the right place for this).
> 
>    dist hbnd, all, all, mode=2
>    hide labels
> 
> (But I don't think that will help with MN.)
> 
> If you don't want dashes, then
> 
>    set dash_length, 3
>    set dash_gap, 0
> 
> Seems to work for me...
> 
> Cheers,
> Warren
> 
> 
> > -----Original Message-----
> > From: pym...@li... 
> > [mailto:pym...@li...] On Behalf Of 
> > John Berrisford
> > Sent: Thursday, February 19, 2004 10:09 AM
> > To: pym...@li...
> > Subject: [PyMOL] drawing hydrogen bonds or dashes
> > 
> > I have searched the manual and forum and cannot find how to 
> > draw a hydrogen bond between two specific atoms. 
> > 
> > I wish to do the equivalent to the following molscript 
> > command "line position res-atom A97 OE1 to position res-atom M200 MN;"
> > 
> > which draws a line between two atoms A97 OE1 and Mn200. So 
> > not a hydrogen bond but a specific line.
> > 
> > Is this possible in pymol?
> > 
> > I have read posts on changing the thickness of a dash and the 
> > number of dashes in a line. But none on how to draw the line.
> > 
> > Can anyone help me, i'm a new user by the way!
> > 
> > thanks
> > 
> > John
> > 
> > 
> > 
> > -------------------------------------------------------
> > SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> > Build and deploy apps & Web services for Linux with a free 
> > DVD software kit from IBM. Click Now!
> > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> > _______________________________________________
> > PyMOL-users mailing list
> > PyM...@li...
> > https://lists.sourceforge.net/lists/listinfo/pymol-users
> > 
> 
> 
> 


-- 
John Berrisford
Department of Molecular Biology and Biotechnology
University of Sheffield

email: j.b...@sh...

Rongjin,

PyMOL can certainly color surfaces by atom color and other properties.

   show surface
   color green, elem c
   color violet, elem n+o 

But Grasp's hydrophobicity function may be doing something more
sophisticated then that...

Cheers,
Warren


> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> gu...@um...
> Sent: Wednesday, February 18, 2004 7:12 PM
> To: pym...@li...
> Subject: [PyMOL] molecular surface colored by hydrophobicity
> 
> Dear all,
> I am preparing a figure of a molecualr surface, colored by 
> the hydrophobicity of surface residues. Can pymol do this job? 
> How about showing atom C* in a color, and N* and O* in another color?
> 
> In fact I have tried GRASP but failed in coloring the 
> surface. I used "r=hyd", but could not color the surface of 
> those redisues. This is an old version of grasp (1.0?). I am 
> not familiar with this program. If anybody can tell me the 
> correct way of doing it in grasp, it will also be greatly appreciated.
> 
> Rongjin Guan
> 
> 
> 
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Morri has a good point:  

	The standalone pymol-0_xx-bin-linux-libc6-i386.tgz binary builds
should be capable of running on virtually any current x86-based Linux
system: RH, SUSE, Debian, Fedora, or whatever.

	So, if you're having trouble with compilation or use of
distribution-specific packages on Linux and simply want to use PyMOL as a
standalone tool, then downloading those binaries is an easy way to get
going.  Installation is trivial and you don't need root privileges.

    1) tar zxvf pymol-(version)-bin-linux-libc6-i386.tgz
    2) cd pymol
    3) ./setup.sh
    4) ./pymol.com 

Cheers,
Warren

FYI: those builds are built using GCC 2.96 under RedHat 6.2 for maximum
compatibility -- they're not the fastest PyMOL binaries in the world, but
they do run without a hassle.
 
> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> Morri Feldman
> Sent: Thursday, February 19, 2004 3:18 PM
> To: Michael Banck
> Cc: pymol users
> Subject: Re: [PyMOL] Ray Tracing Bug in Pymol 0.93 on SuSE Linux 8.2
> 
> Hi Michael,
> 
> I downloaded pymol-0_93-bin-linux-libc6-i386.tgz to my Debian machine.
> When I run this copy rather than the .deb version I am able 
> to ray trace lines without a crash.
> 
> Best,
> Morri
> 
>  On Thu, 19 Feb 2004, Michael Banck wrote:
> 
> > On Thu, Feb 19, 2004 at 08:22:14AM +0100, Thomas Siegmund wrote:
> > > For me PyMol on SuSE 8.2 crashed always when raytracing a 
> line element.
> >
> > That seems to be identical to Debian Bug #229080
> > (http://bugs.debian.org/229080)
> >
> > > The stock  compiler on SuSE 8.2 is gcc 3.3 20030226 
> (prerelease). I 
> > > got the same crash with  gcc 3.3.1, but not with gcc 2.95. I have 
> > > not tried it on SUSE 9 yet, installing and using gcc 2.95 
> (for PyMol 
> > > only) solved the problem for me. See PyMol bugreport 806361.
> >
> > Thanks for the information, I will see whether that helps 
> on Debian, 
> > too.
> >
> >
> > Michael
> >
> >
> > -------------------------------------------------------
> > SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> > Build and deploy apps & Web services for Linux with a free DVD 
> > software kit from IBM. Click Now!
> > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> > _______________________________________________
> > PyMOL-users mailing list
> > PyM...@li...
> > https://lists.sourceforge.net/lists/listinfo/pymol-users
> >
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Matt,

Try:

	set selection_width, 6
	set selection_width, 7
      set selection_width, 8

Cheers,
Warren
 

> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> Matthew Zimmerman
> Sent: Thursday, February 19, 2004 2:46 PM
> To: pym...@li...
> Subject: [PyMOL] Making the selection indicators bigger
> 
> Hello --
> 
> So I've installed PyMOL onto a new system, and I'm having 
> trouble with the little pink indicators that pop up when you 
> make a selection. On this machine they appear to about a 
> single pixel wide, which on a
> 1280x1024 resolution screen makes them practically invisible! 
> I've tried poking about in the options a little but I don't 
> know what to change... 
> Any guidance?
> 
> -- 
>        Matt
> 
>        Matthew Zimmerman
>        Interdisciplinary Biophysics, University of Virginia
>        http://www.people.virginia.edu/~mdz4c/
>        ----------------------------------------------------
>   Lab: 4-223 Jordan Hall, 1300 Jefferson Park Avenue,
>        Charlottesville, VA 22908              (434)924-2948
> Home: 2655-C Barracks Road
>        Charlottesville, VA 22901              (434)244-7989
> 
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with
> a free DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Hi Michael,

I downloaded pymol-0_93-bin-linux-libc6-i386.tgz to my Debian machine.
When I run this copy rather than the .deb version I am able to ray trace
lines without a crash.

Best,
Morri

 On Thu, 19 Feb 2004, Michael Banck wrote:

> On Thu, Feb 19, 2004 at 08:22:14AM +0100, Thomas Siegmund wrote:
> > For me PyMol on SuSE 8.2 crashed always when raytracing a line element.
>
> That seems to be identical to Debian Bug #229080
> (http://bugs.debian.org/229080)
>
> > The stock  compiler on SuSE 8.2 is gcc 3.3 20030226 (prerelease). I
> > got the same crash with  gcc 3.3.1, but not with gcc 2.95. I have not
> > tried it on SUSE 9 yet, installing and using gcc 2.95 (for PyMol only)
> > solved the problem for me. See PyMol bugreport 806361.
>
> Thanks for the information, I will see whether that helps on Debian,
> too.
>
>
> Michael
>
>
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with
> a free DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>

Alex
try typing:

dss


this will assign secondary structure to your PDB and then the cartoon 
will show you the sheets and helices you desire.
Scott


On Feb 19, 2004, at 1:44 PM, ALEX DAJKOVIC wrote:

> dear all,
> i am new to pymol and i have a question about cartoons.
> i saw the demo of cartoon ribbons and really liked it, but i haven't
> been able to get my protein displayed that way.  when i give the 
> command
> "cartoon automatic" all i get is a simple line that apparently traces
> the backbone of my protein chain.  cartoon loop does the same thing.
> with "cartoon tube" i get a thicker line.  how do i set the parameters
> so that i get nice helices and sheets like the ones showed in the demo?
> thanks,
> alex
>
>
>
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with
> a free DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>
>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   Scott Classen, Ph.D.
   ACS Postdoctoral Fellow
   Department of Molecular & Cell Biology
   University of California, Berkeley
   237 Hildebrand Hall #3206
   Berkeley, CA 94720-3206
   LAB 510.643.9491
   FAX 510.643.9290
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Hello --

So I've installed PyMOL onto a new system, and I'm having trouble with 
the little pink indicators that pop up when you make a selection. On 
this machine they appear to about a single pixel wide, which on a 
1280x1024 resolution screen makes them practically invisible! I've tried 
poking about in the options a little but I don't know what to change... 
Any guidance?

-- 
       Matt

       Matthew Zimmerman
       Interdisciplinary Biophysics, University of Virginia
       http://www.people.virginia.edu/~mdz4c/
       ----------------------------------------------------
  Lab: 4-223 Jordan Hall, 1300 Jefferson Park Avenue,
       Charlottesville, VA 22908              (434)924-2948
Home: 2655-C Barracks Road
       Charlottesville, VA 22901              (434)244-7989

dear all, 
i am new to pymol and i have a question about cartoons. 
i saw the demo of cartoon ribbons and really liked it, but i haven't
been able to get my protein displayed that way.  when i give the command
"cartoon automatic" all i get is a simple line that apparently traces
the backbone of my protein chain.  cartoon loop does the same thing. 
with "cartoon tube" i get a thicker line.  how do i set the parameters
so that i get nice helices and sheets like the ones showed in the demo?
thanks, 
alex

John,

Welcome to PyMOL!

> I wish to do the equivalent to the following molscript 
> command "line position res-atom A97 OE1 to position res-atom M200 MN;"

   dist my_line, a/97/OE1, M/200/MN
   hide labels

Also note that PyMOL can draw hydrogen-bonds automatically using its limited
chemical intuition according to the global "h_bond_..." settings. (Sorry
guys, that isn't documented -- I am thinking about creating a new command
for it since "dist" isn't the right place for this).

   dist hbnd, all, all, mode=2
   hide labels

(But I don't think that will help with MN.)

If you don't want dashes, then

   set dash_length, 3
   set dash_gap, 0

Seems to work for me...

Cheers,
Warren


> -----Original Message-----
> From: pym...@li... 
> [mailto:pym...@li...] On Behalf Of 
> John Berrisford
> Sent: Thursday, February 19, 2004 10:09 AM
> To: pym...@li...
> Subject: [PyMOL] drawing hydrogen bonds or dashes
> 
> I have searched the manual and forum and cannot find how to 
> draw a hydrogen bond between two specific atoms. 
> 
> I wish to do the equivalent to the following molscript 
> command "line position res-atom A97 OE1 to position res-atom M200 MN;"
> 
> which draws a line between two atoms A97 OE1 and Mn200. So 
> not a hydrogen bond but a specific line.
> 
> Is this possible in pymol?
> 
> I have read posts on changing the thickness of a dash and the 
> number of dashes in a line. But none on how to draw the line.
> 
> Can anyone help me, i'm a new user by the way!
> 
> thanks
> 
> John
> 
> 
> 
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free 
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

Dan,

	I think it's fair to say that PyMOL's currently molecular building
capabilities are at the stage of a technology
demonstration/proof-of-concept, but very little effort has yet been =
expended
toward making them useful for real work. =20

	Part of that is because we don't yet have a general back-end
modeling engine inside of PyMOL (molecular sculpting is a pure "GIGO" =
kind
of thing).  That is still an important long-term goal.

	But the other reason is that I don't really have a clear idea in my
mind for how such a builder should operate.  What I would like to get =
from
you and all other interested PyMOL users are wish-lists for how PyMOL's
molecular builder should work (better).  No hurry though -- feedback =
over
the next month or two should be fine, and anyone attending MUG or CUP is
welcome to give me an ear-full next week.

	Basically, a major goal is to evolve PyMOL into a builder that can
be useful across the board for biology and drug discovery:  for =
proteins,
small molecules, nucleic acids, carbohydrates, solvent, and so on.  As a
general tool, it should be capable of supporting =
crystallography-oriented
building, manual homology modeling, and hypothetical "hand modeling" of
small molecules, as well as simple illustrative tasks such as the 're
working on.  Of course, specialty tools like "O", "Quanta", "Modeller", =
and
"MacroModel" will always be superior in their respective niches, but I =
hope
PyMOL can become eventually become "good enough" for simple common =
molecular
building tasks...

Questions to consider:

	What are the best builders out there today that we should learn
from?=20
	What are the key functions PyMOL must absolutely have to be a
builder?
	What other extra attractive feature should PyMOL seek to emulate?
	How can we make the best use of what is already in PyMOL, unique or
otherwise?
	What are the biggest "holes" in the current versions?

Things that are already on the radar:

	Rotamer-libraries.
	A 1-D sequence viewer with alignment and editing capabilities.
	An integrated builder tutorial.
	Finding an existing force-field provider to work with...
	Back and forth 2D to 3D for small molecules.

Dan, =20

	It sounds like a feature you would like is that ability to aim the
chain being built towards a certain direction.

	In the meantime, to accomplish your task, build the chain first,
then turn on Auto-Sculpting in the Sculpting menu, put the mouse into
editing mode, then control-left-click-and-drag to move the chain where =
you
want it to go.  Ctrl-middle-click to pick one (or more -- in the latests
betas) joints...and then shift-left-or-middle click-and-drag, or
control-left-click-and-drag to move the resulting fragments.  If your =
chain
is small, increase the number of Cycles per Update to accelerate
convergence.

Cheers,
Warren



> -----Original Message-----
> From: pym...@li...=20
> [mailto:pym...@li...] On Behalf Of=20
> Moriarty, Daniel
> Sent: Thursday, February 19, 2004 9:47 AM
> To: pym...@li...
> Subject: [PyMOL] using the build function
>=20
> Hello everyone,
>=20
> I'm trying to build some model proteins for a diagram showing=20
> a protein unfolding over several steps.  I'm using the build=20
> function in the 0.95beta10 release.  The main problem is that=20
> I can't control the angle of the protein chain, so the chain=20
> just goes off in one direction only, unless proline is placed=20
> in the middle of a helix at which point it goes off at a new=20
> angle.   This makes me think that it can be specified, but I=20
> just don't know how to do it.
>=20
>=20
> Any help would be appreciated,
>=20
> Dan
>=20
>=20
>=20
> -------------------------------------------------------
> SF.Net is sponsored by: Speed Start Your Linux Apps Now.
> Build and deploy apps & Web services for Linux with a free=20
> DVD software kit from IBM. Click Now!
> http://ads.osdn.com/?ad_id=1356&alloc_id438&op=3Dick
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>=20

I have searched the manual and forum and cannot find how to draw a hydrogen bond
between two specific atoms. 

I wish to do the equivalent to the following molscript command
"line position res-atom A97 OE1 to position res-atom M200 MN;"

which draws a line between two atoms A97 OE1 and Mn200. So not a hydrogen bond
but a specific line.

Is this possible in pymol?

I have read posts on changing the thickness of a dash and the number of dashes
in a line. But none on how to draw the line.

Can anyone help me, i'm a new user by the way!

thanks

John

Hello everyone,

I'm trying to build some model proteins for a diagram showing a protein =
unfolding over several steps.  I'm using the build function in the =
0.95beta10 release.  The main problem is that I can't control the angle =
of the protein chain, so the chain just goes off in one direction only, =
unless proline is placed in the middle of a helix at which point it goes =
off at a new angle.   This makes me think that it can be specified, but =
I just don't know how to do it.


Any help would be appreciated,

Dan

* Tsjerk Wassenaar <t.a...@ch...> [2004-02-19 04:30] wrote:
>=20
> Hi Guys,
>=20
> I seem to have some trouble rendering with povray. The render starts and=
=20
> finishes fine, and the image is created, but Pymol crashes with a=20
> segmentation fault as soon as the rendering is done. The image is not=20
> loaded to the viewer. I'm using povray 3.5 with Pymol 0.93 on a laptop=20
> (Intel Centrino) with SuSe 9.0. I have changed povray_exe in povray.py=20
> in modules to "povray".

I never use the povray rendering directly from within PyMOL.  I find the
PyMOL ray-tracer is great for most things and if it isn't sufficient, I
want to be able to edit the povray input as you suggest below.

> Any hints?

Yes!

> Than to Warren probably, it would be great if Pymol would be able to=20
> write a povray object (include file) from a given selection. That would=
=20
> combine the strong points of pymol where it comes to showing the thing=20
> as you want it, with those of Povray, where it comes to lighting,=20
> positioning, colouring, textures and so on. Since Pymol is already=20
> capable of producing povray input it wouldn't be very hard I would guess?

It can do this. I use a simple little script that I call make_pov.py:

######################################
# make_pov.py
=66rom pymol import cmd

def make_pov(file):
  (header,data) =3D cmd.get_povray()
  povfile=3Dopen(file,'w')
  povfile.write(header)
  povfile.write(data)
  povfile.close()
######################################

Load the file with 'run make_pov.py' and then within pymol do:

make_pov('povray.input')

Then outside PyMOL feed this file to povray with whatever command-line
options you need. You do need to look at the size of the image reported
by the make_pov.  It looks something like:

  PyMOL>make_pov('povray.input')
   RayRenderPOV: w 1100 h 900 f   63.349 b  102.310
   RayRenderPOV: vol  -17.316   17.316  -14.168
   RayRenderPOV: vol   14.168   63.349  102.310
   RayRenderPovRay: processed 714 graphics primitives.
   Ray: total time: 0.03 sec. =3D 123885.9 frames/hour. (0.03 sec. accum.)

You'll need to use the same width and height (or the same ratio) to
make your povray output have the correct width to height ratio, e.g.:

  povray +Ipovray.input +W550 +H450

Cheers,
Rob
--=20
Robert L. Campbell, Ph.D.                         <rl...@po...>
Senior Research Associate                            phone: 613-533-6821
Dept. of Biochemistry, Queen's University,             fax: 613-533-2497
Kingston, ON K7L 3N6  Canada       http://adelie.biochem.queensu.ca/~rlc
    PGP Fingerprint: 9B49 3D3F A489 05DC B35C  8E33 F238 A8F5 F635 C0E2

Michael,

* Michael Banck <mb...@gm...> [2004-02-19 12:24] wrote:
> On Thu, Feb 19, 2004 at 08:22:14AM +0100, Thomas Siegmund wrote:
> > For me PyMol on SuSE 8.2 crashed always when raytracing a line element. 
> 
> That seems to be identical to Debian Bug #229080
> (http://bugs.debian.org/229080)
> 
> > The stock  compiler on SuSE 8.2 is gcc 3.3 20030226 (prerelease). I
> > got the same crash with  gcc 3.3.1, but not with gcc 2.95. I have not
> > tried it on SUSE 9 yet, installing and using gcc 2.95 (for PyMol only)
> > solved the problem for me. See PyMol bugreport 806361.
> 
> Thanks for the information, I will see whether that helps on Debian,
> too.

Just as another data point. I use the CVS version of PyMOL (currently
version 0.94) under Debian testing (using gcc-3.3.3 20040125). I haven't
seen any crashes.  (In case it makes a difference: Athlon XP+ 2500 and
Radeon 9500 Pro graphics).

Cheers,
Rob
-- 
Robert L. Campbell, Ph.D.                         <rl...@po...>
Senior Research Associate                            phone: 613-533-6821
Dept. of Biochemistry, Queen's University,             fax: 613-533-2497
Kingston, ON K7L 3N6  Canada       http://adelie.biochem.queensu.ca/~rlc
    PGP Fingerprint: 9B49 3D3F A489 05DC B35C  8E33 F238 A8F5 F635 C0E2