pymol-users — Collaborative support and tips for PyMOL users

RE: [PyMOL] Linux install problem

[DeLano, W. <wa...@su...>]

Gareth,

   That is an unusual error.  I suspect it has something to do with your =
symbolic links, which isn't the most reliable way to solve the =
dependency problem.  Instead, delete those symlinks, and modify =
Rules.make to seek out the system libraries in the usual location.  =
Rules.linux-rh7x should be your starting point for Rules.make.

Cheers,
Warren

--
mailto:wa...@su...
Warren L. DeLano, Ph.D.
Informatics Manager
Sunesis Pharmaceuticals, Inc.
341 Oyster Point Blvd.
S. San Francisco, CA 94080
(650)-266-3606  FAX:(650)-266-3501



> -----Original Message-----
> From: Gareth Stockwell [mailto:ga...@eb...]
> Sent: Friday, May 03, 2002 4:00 AM
> To: pym...@li...
> Subject: [PyMOL] Linux install problem
>=20
>=20
>=20
> I am having a problem installing PyMol 0.80 on my (Redhat=20
> 7.1) Linux box.
>=20
> I already have Tcl, Tk, Python 2.1, zlib and libpng so I just=20
> made libglut=20
> from the ext-0.79 package.
>=20
>   cd $PYMOL_PATH/ext
>   ./build.com glut-linux
>=20
> The GL libraries are in strange places on my system, so I made the=20
> following links:
>=20
>   ln -s /usr/lib/libGL.so.1 $PYMOL_PATH/ext/lib/libGL.so
>   ln -s /usr/lib/libGLU.so.1 $PYMOL_PATH/ext/lib/libGLU.so
>   ln -s /usr/X11R6/lib/modules/extensions/libGLcore.a \
>      $PYMOL_PATH/ext/lib/libGLcore.a
>=20
> Then I built the PyMol program
>=20
>   cd $PYMOL_PATH/pymol
>   vi Rules.make
>   make
>=20
> The 'make' succeeds, but when I execute pymol.com, I get
>=20
> Traceback (most recent call last):
>   File=20
> "/ebi/research/thornton/ligands/gareth/src/pymol-linux/pymol/m
> odules/launch_pymol.py",=20
> line 19, in ?
>     import pymol
>   File=20
> "/ebi/research/thornton/ligands/gareth/src/pymol-linux/pymol/m
> odules/pymol/__init__.py",=20
> line 137, in ?
>     import _cmd
> ImportError: /usr/X11R6/lib/libICE.so.6: shared object not open
>=20
>=20
> This shared object file exists - is there any reason why this=20
> error should=20
> occur?
>=20
> Gareth
>=20
> P.S. I don't have root access on this system
>=20
> --------------------------------------------------------------
> -----------------
>  Gareth Stockwell
>  PhD student, Thornton Group
>  EMBL - European Bioinformatics Institute
>  Wellcome Trust Genome Campus
>  Hinxton
>  Cambridge CB10 1SD                                        =20
> ga...@eb...
>  Tel 01223 492548                              =20
> http://www.ebi.ac.uk/~gareth
>=20

[PyMOL] Linux install problem

[Gareth S. <ga...@eb...>]

I am having a problem installing PyMol 0.80 on my (Redhat 7.1) Linux box.

I already have Tcl, Tk, Python 2.1, zlib and libpng so I just made libglut 
from the ext-0.79 package.

  cd $PYMOL_PATH/ext
  ./build.com glut-linux

The GL libraries are in strange places on my system, so I made the 
following links:

  ln -s /usr/lib/libGL.so.1 $PYMOL_PATH/ext/lib/libGL.so
  ln -s /usr/lib/libGLU.so.1 $PYMOL_PATH/ext/lib/libGLU.so
  ln -s /usr/X11R6/lib/modules/extensions/libGLcore.a \
     $PYMOL_PATH/ext/lib/libGLcore.a

Then I built the PyMol program

  cd $PYMOL_PATH/pymol
  vi Rules.make
  make

The 'make' succeeds, but when I execute pymol.com, I get

Traceback (most recent call last):
  File 
"/ebi/research/thornton/ligands/gareth/src/pymol-linux/pymol/modules/launch_pymol.py", 
line 19, in ?
    import pymol
  File 
"/ebi/research/thornton/ligands/gareth/src/pymol-linux/pymol/modules/pymol/__init__.py", 
line 137, in ?
    import _cmd
ImportError: /usr/X11R6/lib/libICE.so.6: shared object not open


This shared object file exists - is there any reason why this error should 
occur?

Gareth

P.S. I don't have root access on this system

-------------------------------------------------------------------------------
 Gareth Stockwell
 PhD student, Thornton Group
 EMBL - European Bioinformatics Institute
 Wellcome Trust Genome Campus
 Hinxton
 Cambridge CB10 1SD                                         ga...@eb...
 Tel 01223 492548                               http://www.ebi.ac.uk/~gareth

[PyMOL] RE: easy (?) question (DeLano, Warren)

[Marc S. <mar...@mp...>]

>    2. Let's Introduce Your Company To China Government (inv...@ya...)
      ^ ^ ^ ^
Shouldn't that have been filtered out?


> So long as the N and C atoms are also present, the ribbon representation 
> can do this.  It looks for CA atoms three bonds apart to form the trace. 
> To get a linear C-alpha trace (no smoothing), "set ribbon_sampling 1"

This is a bit unobvious (at least for me) wouldn't it be good to have 
this as an extra menu-point (i.. CA-trace)?

-- 
Bye,  Marc Saric

[PyMOL] Fast rendering option?

[Gareth S. <ga...@eb...>]

Is there a 'fast rendering' option in PyMol, as found in other molecular 
graphics suites?

By this I mean that, when using a 'slow-rendering' view such as spheres,
is there a way to make PyMol switch to 'lines only' when the mouse button
is depressed, so that rotations are fast, then re-render the spheres when
motion stops?

Gareth

-------------------------------------------------------------------------------
 Gareth Stockwell
 PhD student, Thornton Group
 EMBL - European Bioinformatics Institute
 Wellcome Trust Genome Campus
 Hinxton
 Cambridge CB10 1SD                                         ga...@eb...
 Tel 01223 492548                               http://www.ebi.ac.uk/~gareth

[PyMOL] Let's Introduce Your Company To China Government

[<inv...@ya...>]

<html>
<head>
<title>investcn.biz</title>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
</head>

<body bgcolor="#FFFFFF" text="#000000">
<table width="575" cellspacing="0" cellpadding="0" align="center">
  <tr> 
    <td bgcolor="#FF0000"> 
      <table width="100%" cellspacing="3" cellpadding="3">
        <tr>
          <td>
            <div align="center"><font face="Arial, Helvetica, sans-serif" size="3"><b><font size="4" color="#FF0000"><font color="#FFFFFF">Let 
              China Government Gets To Know Your Company</font></font></b></font></div>
          </td>
        </tr>
      </table>
      
    </td>
  </tr>
  <tr>
    <td>
      <div align="center"><font face="Arial, Helvetica, sans-serif" size="3"><b><font size="4" color="#FF0000"><u><font size="3" color="#0000FF">The 
        One and Only B2G Business Directory <br>
        (in Simplied Chinese) for the China Government Officials!</font></u></font></b></font></div>
    </td>
  </tr>
  <tr> 
    <td> 
      <div align="center"><font face="Arial, Helvetica, sans-serif" size="2"><b><font size="3"><img src="http://us.f1.yahoofs.com/users/1ac3b719/bc/b2g/2_books_cover_00.jpg?bc1OM08AuPCm9_Sr" width="300" height="236"></font></b></font></div>
    </td>
  </tr>
  <tr> 
    <td> 
      <div align="center"><font face="Arial, Helvetica, sans-serif" size="2"><b><font size="3">First 
        500 advertisers will get free handbook<br>
        <font color="#0000FF">The 1000 Billion Purchasing Market Handbook</font> 
        <br>
        <font size="2" color="#FF0000">( How to access China Government Purchasing 
        Market )</font></font></b></font></div>
    </td>
  </tr>
  <tr> 
    <td>
      <p align=justify><font size="2" face="Arial, Helvetica, sans-serif"><i>Invest</i><font face="Times New Roman, Times, serif"><b>CN</b></font> 
        is pleased to announce that its first China B2G Directory will be published 
        in Oct 2002. This B2G Directory is not for sale. First edition will be 
        published 5000 issues and distributed to : State Development Planning 
        Commission System, Ministries and Commissions under the State Council, 
        Chief Executive Bureau of all Provinces, Cities, and Counties (more than 
        2000 counties), Organizations directly under the State Council, and Institutions 
        directly under the State Council. </font> 
      <p align="center"><font size="2" face="Arial, Helvetica, sans-serif"><b><font color="#009900" size="3">If 
        you are interested to do business with China government, <br>
        don't miss out this opportunity. <br>
        Advertising and listing spaces will run out fast! First come first served!</font></b></font></p>
      <p> 
      <p align=justify><font size="2" face="Arial, Helvetica, sans-serif">By way 
        of background, <i>Invest</i><font face="Times New Roman, Times, serif"><b>CN</b></font>, 
        is a division of China Economic Herald, which is the media bureau of State 
        Development Planning Commission, P.R.China. The main objective of <i>Invest</i><font face="Times New Roman, Times, serif"><b>CN</b></font> 
        is to provide practical, deliverable, workable China investment solutions 
        to investors, bankers, CEOs, CFOs, COOs, decision-making officers, purchasing 
        managers, government officials at all levels around the world.</font></p>
      <p align=center><b><font size="2" face="Arial, Helvetica, sans-serif">To 
        reserve your listing on our China B2G Directory, email <b><a href="mailto:inv...@ya...">inv...@ya...</a></b></font></b></p>
    </td>
  </tr>
  <tr> 
    <td> 
      <hr>
    </td>
  </tr>
  <tr> 
    <td> 
      <div align="center"><font size="2" face="Arial, Helvetica, sans-serif"><b>THIS 
        IS NOT SPAM, YOU ARE RECEIVING THIS INFORMATION BECAUSE :</b><br>
        A) You submitted a classified ad or link, posted your e-mail address in 
        public commercial domains.<br>
        B) You sent email to one of our domains and/or addresses in the past or<br>
        C) We are on the same list. If you were not the intended recipient of 
        this <br>
        message, Please accept our apologies and delete. If you wish to no longer 
        <br>
        receive email from us please send an email message to <b><a href="mailto:inv...@ya...">inv...@ya...</a></b><br>
        with &quot;Remove&quot; placed in the subject line. Thank you.</font></div>
    </td>
  </tr>
  <tr> 
    <td> 
      <table width="100%" cellspacing="0" cellpadding="0">
        <tr> 
          <td> 
            <hr>
          </td>
          <td width="20%"> 
            <div align="center"><font size="2"><b><font face="Arial, Helvetica, sans-serif">DISCLAIMER</font></b></font></div>
          </td>
          <td> 
            <hr>
          </td>
        </tr>
      </table>
    </td>
  </tr>
</table>
<br>
</body>
</html>

RE: [PyMOL] easy (?) question

[DeLano, W. <wa...@su...>]

> From: Byron DeLaBarre [mailto:by...@SL...]
>=20
> Anyone out there know how to prevent bonding of adjacent but=20
> chemically
> unconnected atoms in pymol?

Two step process:

1. edit the PDB file and replace "ATOM  " with "HETATM"
2. after the HETATM's add a single record "CONECT" following by nothing =
but a new-line.  This will signal PyMOL to expect explicit connectivity =
(of which there will be none).  In the future, if you need to specify =
connectivity explicitly, this is how you do it using a PDB file (CONECT =
records followed by connectivity info)

Note if you need to surface hetatms, "set surface_mode=3D1" or "alter =
(all),type=3D'ATOM'" followed by "sort" and "rebuild".  Same for goes =
for meshes.

> And along those lines, anyone had success with showing a=20
> Calpha trace?  I
> tried once but couldn't figure out how to get pymol to connect the
> individual Calpha atoms.  I think this must be easy to do =20
> but I couldn't
> work it out at the time.

So long as the N and C atoms are also present, the ribbon representation =
can do this.  It looks for CA atoms three bonds apart to form the trace. =
  To get a linear C-alpha trace (no smoothing), "set =
ribbon_sampling=3D1"

Warren
=20

[PyMOL] easy (?) question

[Byron D. <by...@SL...>]

Hi Pymolers -

Anyone out there know how to prevent bonding of adjacent but chemically
unconnected atoms in pymol?

It is possible to go in and unconnected bonds that are nonsense, but that is
kind of a pain to do every time.

And along those lines, anyone had success with showing a Calpha trace?  I
tried once but couldn't figure out how to get pymol to connect the
individual Calpha atoms.  I think this must be easy to do  but I couldn't
work it out at the time.

I tried playing with the bonding_vdw_cutoff setoff (my best guess for how to
fix both problems) but didn't see anything happening.


Thanks -

Byron

Byron DeLaBarre, Ph. D.
P250 MSLS Building
Stanford University
(650)736-1714

Re: [PyMOL] Re: image quality

[Chris R. <chr...@va...>]

Hi,

Thanks to Andrey and Warren, that was certainly what was needed!

Chris

On Tuesday, April 30, 2002, at 12:03  PM, DeLano, Warren wrote:

> Image quality:
>
>    As Andrey indicated, on typical color lasers and inkjets, you need 
> 300 dots per final printed inch (~120 pixels/cm) for maximum quality .  
> A small 4"x3" illustration would need to be 1200x900 pixels ("ray 
> 1200,900").  Full page 11"x8.5" (ray "3300,2550").
>
> Cheers,
> Warren
>> Create image with high (I mean really High) resolution.
>>
>> Printers have much better resolution (at least 300dpi) and to
>> print your
>> low-resolution screen image, they have to scale it.  So, the grains...
>>
>> --
>> Andrey V Khavryuchenko          http://www.kds.com.ua/
>> Offshore Software Development
>

RE: [PyMOL] Re: image quality

[DeLano, W. <wa...@su...>]

Image quality:

   As Andrey indicated, on typical color lasers and inkjets, you need =
300 dots per final printed inch (~120 pixels/cm) for maximum quality .  =
A small 4"x3" illustration would need to be 1200x900 pixels ("ray =
1200,900").  Full page 11"x8.5" (ray "3300,2550").  =20

Since rendering those images takes forever and the files get really =
huge, I tend to trade quality for time and space, and stick with 150-200 =
pixels per inch (~60-75 pixels/cm) for draft/in-house printouts.

Cheers,
Warren

-
mailto:wa...@su...
Warren L. DeLano, Ph.D.
Informatics Manager
Sunesis Pharmaceuticals, Inc.
341 Oyster Point Blvd.
S. San Francisco, CA 94080
(650)-266-3606  FAX:(650)-266-3501



> -----Original Message-----
> From: Andrey Khavryuchenko [mailto:ak...@kd...]
> Sent: Tuesday, April 30, 2002 9:30 AM
> To: Chris Rife
> Cc: pym...@li...
> Subject: [PyMOL] Re: image quality
>=20
>=20
> Chris,
>=20
> "CR" =3D=3D Chris Rife wrote:
>=20
>  CR> I'm using Pymol to make some images for a paper and a=20
> poster, and I've
>  CR> run across a problem. I can generate images that are=20
> beautiful on my
>  CR> screen (when ray traced and then viewed as the png=20
> file), but when I
>  CR> print them out they become extremely grainy. I've tried=20
> viewing and
>  CR> printing from different programs, but with no luck. Any=20
> suggestions?
>=20
> Create image with high (I mean really High) resolution. =20
>=20
> Printers have much better resolution (at least 300dpi) and to=20
> print your
> low-resolution screen image, they have to scale it.  So, the grains...
>=20
> --=20
> Andrey V Khavryuchenko          http://www.kds.com.ua/
> Offshore Software Development
>=20
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>=20

[PyMOL] Re: image quality

[Andrey K. <ak...@kd...>]

Chris,

"CR" == Chris Rife wrote:

 CR> I'm using Pymol to make some images for a paper and a poster, and I've
 CR> run across a problem. I can generate images that are beautiful on my
 CR> screen (when ray traced and then viewed as the png file), but when I
 CR> print them out they become extremely grainy. I've tried viewing and
 CR> printing from different programs, but with no luck. Any suggestions?

Create image with high (I mean really High) resolution.  

Printers have much better resolution (at least 300dpi) and to print your
low-resolution screen image, they have to scale it.  So, the grains...

-- 
Andrey V Khavryuchenko          http://www.kds.com.ua/
Offshore Software Development

[PyMOL] image quality

[Chris R. <chr...@va...>]

Hi,

I'm using Pymol to make some images for a paper and a poster, and I've 
run across a problem. I can generate images that are beautiful on my 
screen (when ray traced and then viewed as the png file), but when I 
print them out they become extremely grainy. I've tried viewing and 
printing from different programs, but with no luck. Any suggestions?

Thanks,
Chris

Re: [PyMOL] win32 build

[Warren L. D. <wa...@su...>]

On Sun, 28 Apr 2002, Trevor Kramer wrote:

> I would like to build pymol for win32 from the source but I 
> cannot find instructions to do so in cvs.  Is this possible?  Thanks.

Yes, of course it is possible, but my current approach is a mess.

At present (April 2002):

I use a combination of cygwin and VC++ to update, build, and test the
windows version.  It involves a collection of project directories
with binary project files, and I end up hard-linking in components which
really shouldn't be hard-linked in (numpy, pyopengl, etc.). Plus, I end
up embedding the Python interpretor on Windows instead of building PyMOL
into a DLL module (as on Unix).  In fact, there is a complete different
launch sequence for PyMOL under Windows than under Unix [though users
shouldn't notice any difference : )]  

In summary, the current win32 build is complex, fragile, and not-Python
like.  It should not be the example for others to follow, and that's
why I don't distribute it (plus, we're talking >100 megabytes to
faithfully reproduce the build environment).  What the PyMOL project
needs is help from someone who is a more Win32-savvy, and who can take the
lead in developing a robust Windows build that is consistent with the
approach taken by other Python-based projects.  

Specifically, I am thinking about command-line-based build, python-like
module DLLs, and a unix-like launch sequence -- without changing the
the user-experience (command line from DOS or cygin, or double-click
launch, in that case with a visible standard-output window).  If anyone
can pull that off, then I'll gladly make their version into the
"standard" win32 build...

Any volunteers?  If not, then this will remain just one more entry on
my long list of things to do.

Although I don't have much advice to offer, here are the build options I
current use under VC++ with a Python-embedded build (in this example,
using Python22)  Also note that PYTHON_PATH should be defined in the
environment before PyMOL is launched.  

/nologo /ML /W2 /GX /O2 /I "." /I "layer0" /I "layer1" /I "layer2" /I
"layer3" /I "layer4" /I "layer5" /I "c:\python22\include" /I
"contrib/modules" /I "contrib/sglite" /D "WIN32" /D "NDEBUG" /D
"_CONSOLE" /D "_MBCS" /D "HAVE_NUMERIC_ARRAYOBJECT_H" /D "_PYMOL_NUMPY" /D
"_HAVE_LIBPNG" /D "_PYMOL_MONOLITHIC" /D
"PythonTypes" /Fp"PyMOL___Win32_Release_Python221/PyMOL.pch" /YX
/Fo"PyMOL___Win32_Release_Python221/" /Fd"PyMOL___Win32_Release_Python221/" /FD
/c

The source-code in CVS is Win32-competent -- all you need is to compile it
and then link against the following libraries (some which can be found on
the internet).  zlib and libpng are straight off the internet.

zlib.lib libpng.lib kernel32.lib user32.lib gdi32.lib winspool.lib
comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib
odbc32.lib odbccp32.lib kernel32.lib user32.lib gdi32.lib
winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib
uuid.lib odbc32.lib odbccp32.lib "c:\python22\libs\python22.lib" /nologo
/subsystem:console /incremental:no
/pdb:"PyMOL___Win32_Release_Python221/PyMOL.pdb" /debug /machine:I386
/nodefaultlib:"msvcrt.lib" /out:"PyMOL.exe"

Cheers, 
Warren

[PyMOL] win32 build

[Trevor K. <tk...@ha...>]

I would like to build pymol for win32 from the source but I cannot find =
instructions to do so in cvs.  Is this possible?  Thanks.

Trevor

RE: [PyMOL] color association with state?

[DeLano, W. <wa...@su...>]

> From: Malcolm E Davis [mailto:mal...@bm...]
>=20
> A follow-up newbie question:  Is it possible to map colors=20
> onto a surface on a per vertex basis?

Not yet, but that functionality is planned for sometime this year in =
order to enable painting of surfaces in a non-atomic fashion.  For now, =
you'll have to make do with atom-based coloring or use something like =
Grasp.

- Warren

=20

Re: [PyMOL] color association with state?

[Malcolm E D. <mal...@bm...>]

"DeLano, Warren" wrote:

> Sorry, this can't be done with the current versions -- colors are either atomic or whole-object properties (with a couple of minor exceptions).

A follow-up newbie question:  Is it possible to map colors onto a surface on a per vertex basis?

Malcolm

[PyMOL] compiling on Solaris

[Joel H. <jh...@un...>]

Dear PyMolers

We have a fellow here who is trying hard to compile
pymol on a Solaris and encountering some problems.
Has anyone tried this and, if so, can you give us
some advice.

Thanks.

Joel

Joel M. Harp, Ph.D.
Department of Biochemistry and Molecular Genetics
University of Virginia Health System
Charlottesville VA 22908-0733
USA

RE: [PyMOL] color association with state?

[DeLano, W. <wa...@su...>]

> From: Byron DeLaBarre [mailto:by...@SL...]
> Subject: [PyMOL] color association with state?

> Do any pymolers out there know how to associate the color of=20
> a single object
> with its state?

Sorry, this can't be done with the current versions -- colors are either =
atomic or whole-object properties (with a couple of minor exceptions).  =
The only way around this right now is to distribute the states you want =
to show over a set of objects with different colors:

isomesh m1,map1,1.0,state=3D1
isomesh m2,map2,1.0,state=3D2
isomesh m3,map3,1.0,state=3D3

color red,m1
color blue,m2
color green,m3

> I have a multi-state map that I want to color as it moves through the
> different states in a movie.  I want the colors to be associated with
> specific states.  I only know how to color the entire=20
> collection of states a
> single color.  (something like: color =3D red, object)

> Alternatively, how could I associate a number of text objects=20
> (you can do
> those in pymol, right?) with individual states within the=20
> collection of
> states.

Text objects are brand new, and can only be built right now using =
compiled graphics objects (CGOs).

CGOs are always associated with specific states, so adding a label which =
changes during the course of a movie would be straightforward, except =
that text objects are completely undocumented.  Here is a start:

# REQUIRES PyMOL 0.80
# save as cgo_3Dtext01.py

from pymol import cmd
from pymol.cgo import *
from pymol.vfont import plain

cgo =3D []

axes =3D [[2.0,0.0,0.0],[0.0,2.0,0.0],[0.0,0.0,2.0]]

pos =3D [0.0,0.0,0.0]
wire_text(cgo,plain,pos,'Hello World',axes)

pos =3D [0.0,-3.0,0.0]
cyl_text(cgo,plain,pos,'Hello Universe',0.10,axes=3Daxes)

cmd.set("cgo_line_radius",0.03)
cmd.load_cgo(cgo,'txt')
cmd.zoom("all",2.0)

- Warren

mailto:wa...@su...
Warren L. DeLano, Ph.D.

[PyMOL] color association with state?

[Byron D. <by...@SL...>]

Do any pymolers out there know how to associate the color of a single object
with its state?

I have a multi-state map that I want to color as it moves through the
different states in a movie.  I want the colors to be associated with
specific states.  I only know how to color the entire collection of states a
single color.  (something like: color = red, object)

Alternatively, how could I associate a number of text objects (you can do
those in pymol, right?) with individual states within the collection of
states.

Thanks -

Byron

Byron DeLaBarre, Ph. D.
P250 MSLS Building
Stanford University
(650)736-1714

re: [PyMOL] Fancy images

[Jacob C. <jc...@gl...>]

Hi Robert,
To do that, you'll need to become moderately well acquainted with
working with a POVRay scene file.

It sounds like what you're thinking of would be covered by the "plane"
object in POVRay. As described in the POVRay manual, planes are
defined as follows

#begin plane definition
plane { <0, 1, 0>, -1
    pigment {color orange}
  }

This defines an infinite plane (great for a background) that's colored
orange. The vector <0,1,0> (in the format <x,y,z>) is the surface normal
of the plane (i.e. if we were standing on the surface, the normal points
straight up along the y axis). The number after the vector definition is the
distance that the plane is displaced along the normal from the
origin -- in this case, the floor is placed at y=-1 so that a sphere
at y=1, radius=2, would be resting on it.

What you'll need to do is have PyMOL output your scene file, figure
out where the boundaries of your molecule are, then displace a plane
away from the origin both far enough that it doesn't clip into the molecule
and in the right direction, such that your light source causes the shadow
of your molecule to fall upon the plane.

I hope I was able to be of some help.

Jacob

  
RC> Hi,

RC> Someone posed a question to me that I couldn't answer, so I'm turning to
RC> the collective wisdom here for help.

RC> How does one create one of those fancy journal-cover images in which,
RC> say, a structure is superimposed on some other image as a background,
RC> but in which a shadow is cast on the background. There is a simple image
RC> of this sort on the opening page of the pymol gallery, so I figure this
RC> must be possible and that perhaps Warren himself knows. :)

RC> I assume that this might be a povray method, so I figured out that if do
RC> something like:

RC>         (header,data) = cmd.get_povray()
RC>         file=open('povray.dat','w')
RC>         file.write(header)
RC>         file.write(data)
RC>         file.close()

RC> then I have a povray input file that I can render.

RC> Does anybody have a recipe for adding a background image using povray?
RC> Or is there another, better way?

RC> Cheers,
RC> Rob...@li...

RE: [PyMOL] Fancy images

[DeLano, W. <wa...@su...>]

Robert,

   You are on the right track.  In order to do this right, you'll need =
to use PovRay, which supports perspective and textures. =20

   PyMOL's raytracer is really just an optimized orthographic, =
textureless, reflectionless raycaster, which is fine for simple =
molecules, but inadequate for complex scenes.  In other words, to make a =
Science or Nature cover image, you will have to do a bit more work than =
just hitting the "Ray" button. =20

   Specifically, you need to introduce a geometric object behind the =
molecule onto which the shadow can be cast.  If all you need is a flat, =
untextured surface, then you can use PyMOL's CGO module to generate =
this.  Otherwise, you'll need to learn how to modify PyMOL's PovRay =
input file to contain these objects by reading the PovRay documentation =
and editing the text file you already know how to generate.

  I've attached an example below.  Save to "ray.py" and run it from =
within PyMOL. =20

Cheers,
Warren


from pymol.cgo import *

obj =3D [
      BEGIN, TRIANGLE_STRIP,
      COLOR, 0.8,0.7,0.4,
      NORMAL, 0.0, 1.0, 0.0,

      VERTEX, -7.0, -20.0, 0.0,
      VERTEX, -10.0, -20.0, 30.0,

      VERTEX,  12.0, -20.0, 0.0,
      VERTEX,  15.0, -20.0, 30.0,

      END,
      ]

cmd.load_cgo(obj,"plane")
cmd.load("$PYMOL_PATH/test/dat/pept.pdb")
util.ray_shadows('heavy')

cmd.set_view((\
     0.962451875,   -0.074250713,   -0.261098653,\
     0.192369312,    0.865197897,    0.463061303,\
     0.191519246,   -0.495900899,    0.846994936,\
    -0.427299917,    0.681541085,  -83.549995422,\
     1.224037170,  -10.279197693,   20.545440674,\
    70.968811035,  117.627342224,    0.000000000 ))

cmd.ray()

--
mailto:wa...@su...
Warren L. DeLano, Ph.D.

> -----Original Message-----
> From: Robert Campbell [mailto:rl...@k2...]
> Sent: Friday, April 12, 2002 9:52 AM
> To: PyM...@li...
> Subject: [PyMOL] Fancy images
>=20
>=20
> Hi,
>=20
> Someone posed a question to me that I couldn't answer, so I'm=20
> turning to
> the collective wisdom here for help.
>=20
> How does one create one of those fancy journal-cover images in which,
> say, a structure is superimposed on some other image as a background,
> but in which a shadow is cast on the background. There is a=20
> simple image
> of this sort on the opening page of the pymol gallery, so I=20
> figure this
> must be possible and that perhaps Warren himself knows. :)
>=20
> I assume that this might be a povray method, so I figured out=20
> that if do
> something like:
>=20
> 	(header,data) =3D cmd.get_povray()
> 	file=3Dopen('povray.dat','w')
> 	file.write(header)
> 	file.write(data)
> 	file.close()
>=20
> then I have a povray input file that I can render.
>=20
> Does anybody have a recipe for adding a background image using povray?
> Or is there another, better way?
>=20
> Cheers,
> Robert
> --=20
> Robert L. Campbell, Ph.D.              =20
> http://biophysics.med.jhmi.edu/rlc
> rl...@k2...                                    phone:=20
> 410-614-6313
> Research Specialist/X-ray Facility Manager
> HHMI/Dept. of Biophysics & Biophysical Chem., The Johns=20
> Hopkins University
>     PGP Fingerprint: 9B49 3D3F A489 05DC B35C  8E33 F238 A8F5=20
> F635 C0E2
>=20
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
>=20

[PyMOL] Fancy images

[Robert C. <rl...@k2...>]

Hi,

Someone posed a question to me that I couldn't answer, so I'm turning to
the collective wisdom here for help.

How does one create one of those fancy journal-cover images in which,
say, a structure is superimposed on some other image as a background,
but in which a shadow is cast on the background. There is a simple image
of this sort on the opening page of the pymol gallery, so I figure this
must be possible and that perhaps Warren himself knows. :)

I assume that this might be a povray method, so I figured out that if do
something like:

	(header,data) = cmd.get_povray()
	file=open('povray.dat','w')
	file.write(header)
	file.write(data)
	file.close()

then I have a povray input file that I can render.

Does anybody have a recipe for adding a background image using povray?
Or is there another, better way?

Cheers,
Robert
-- 
Robert L. Campbell, Ph.D.               http://biophysics.med.jhmi.edu/rlc
rl...@k2...                                    phone: 410-614-6313
Research Specialist/X-ray Facility Manager
HHMI/Dept. of Biophysics & Biophysical Chem., The Johns Hopkins University
    PGP Fingerprint: 9B49 3D3F A489 05DC B35C  8E33 F238 A8F5 F635 C0E2

RE: [PyMOL] atom picking on windows

[Sherwood, P (Paul) <P.S...@dl...>]

Warren

Yes ... this did the trick. The machine display was set to 65536 colours
so I switched to 16777216.

thanks!

Paul

-----Original Message-----
From: DeLano, Warren [mailto:wa...@su...]
Sent: 11 April 2002 17:46
To: Paul Sherwood
Subject: RE: [PyMOL] atom picking on windows


Paul,

   This can sometimes happen on PC graphics cards with insufficient colors.
Are you running in 16, 24, or 32 bit mode?  Try switching to a higher bit
depth if you can...

Please let me know if this solves your problem.

- Warren

PS    There is a different problem with picking on some versions of IRIX,
which is related to a hardware driver problem which affects mouse motion
while keyboard modifiers are held down.  This is mentioned in the PyMOL FAQ.

--
mailto:wa...@su...
Warren L. DeLano, Ph.D.
Informatics Manager
Sunesis Pharmaceuticals, Inc.
341 Oyster Point Blvd.
S. San Francisco, CA 94080
(650)-266-3606  FAX:(650)-266-3501



> -----Original Message-----
> From: Paul Sherwood [mailto:p.s...@dl...]
> Sent: Thursday, April 11, 2002 5:53 AM
> To: PyM...@li...
> Subject: [PyMOL] atom picking on windows
> 
> 
> Dear Warren and fellow users,
> 
> I am having trouble with atom picking on windows NT, using 
> pymol 0.80 (with python 2.1).
> 
> If I start from the methane fragment I find it hard to pick
> the H atoms, while I can select them all I have to click
> some way away from the atom to select it (actually I find I 
> need to click close to the bond between C and another H atom)
> 
> BTW I have one of the MS mice which has the middle mouse replaced
> by a wheel.. mildly inconvenient for pickling with, but as far as
> I can see this is not the problem. 
> 
> Similar picking works on my SGI using a recent CVS checkout,
> (a month or so old, sorry I don't have an exact checkout date). 
> I have noticed similar situations arising on the SGI in the past
> which trying to set up the (lb) selection, but they are not so
> consistent and I cannot reproduce them now.
> 
> Any ideas?
> 
> thanks in advance, 
> 
> Paul
> 
> -- 
> ======================================================================
> Dr Paul Sherwood,         email:  p.s...@da...
> CLRC Daresbury Lab,       tel:    +44-1925-603553
> Warrington,               fax:    +44-1925-603634
> WA4 4AD,  UK.             http:   http://www.cse.clrc.ac.uk
> ======================================================================
> 
> _______________________________________________
> PyMOL-users mailing list
> PyM...@li...
> https://lists.sourceforge.net/lists/listinfo/pymol-users
> 

RE: [PyMOL] atom picking on windows

[Sherwood, P (Paul) <P.S...@dl...>]

Jacob

thanks for your help .. I hadn't noticed the 2 button mode.
But the shift-left-mouse now behaves the same
as the ctrl-middle-mouse did before, namely selecting an atom
other than the one the mouse is over.

Any more ideas?

thanks again,

Paul

-----Original Message-----
From: Jacob Corn [mailto:jc...@gl...]
Sent: 11 April 2002 16:55
To: PyM...@li...
Subject: Re: [PyMOL] atom picking on windows


Paul,

I've had the same problem on both Win98 and Win2k using PyMOL 
0.74-0.80, but with a Logitech wheel mouse. The easiest solution that 
I've found is to switch the mouse layout to 2-button, rather than 
3-button, and just use Shift-click (if I remember correctly) to pick 
atoms.
If you're very used to the wheel-click to pick, you could also remap 
the wheel button to Shift-Click, and from there switching to 
2-button layout will allow you to click the scroll wheel to pick 
atoms. You won't get the benefits of the full 3-button layout, but 
it'll save you having to hold a key down every time you want to pick 
an atom.

Jacob

>Dear Warren and fellow users,
>
>I am having trouble with atom picking on windows NT, using
>pymol 0.80 (with python 2.1).
>
>If I start from the methane fragment I find it hard to pick
>the H atoms, while I can select them all I have to click
>some way away from the atom to select it (actually I find I
>need to click close to the bond between C and another H atom)
>
>BTW I have one of the MS mice which has the middle mouse replaced
>by a wheel.. mildly inconvenient for pickling with, but as far as
>I can see this is not the problem.
>
>Similar picking works on my SGI using a recent CVS checkout,
>(a month or so old, sorry I don't have an exact checkout date).
>I have noticed similar situations arising on the SGI in the past
>which trying to set up the (lb) selection, but they are not so
>consistent and I cannot reproduce them now.
>
>Any ideas?
>
>thanks in advance,
>
>Paul
>
>--
>======================================================================
>Dr Paul Sherwood,         email:  p.s...@da...
>CLRC Daresbury Lab,       tel:    +44-1925-603553
>Warrington,               fax:    +44-1925-603634
>WA4 4AD,  UK.             http:   http://www.cse.clrc.ac.uk
>======================================================================
>
>_______________________________________________
>PyMOL-users mailing list
>PyM...@li...
>https://lists.sourceforge.net/lists/listinfo/pymol-users

-- 

Jacob Corn
Gladstone Institute of Neurological Disease
415.695.3745
jc...@gl...

_______________________________________________
PyMOL-users mailing list
PyM...@li...
https://lists.sourceforge.net/lists/listinfo/pymol-users

Re: [PyMOL] atom picking on windows

[Jacob C. <jc...@gl...>]

Paul,

I've had the same problem on both Win98 and Win2k using PyMOL 
0.74-0.80, but with a Logitech wheel mouse. The easiest solution that 
I've found is to switch the mouse layout to 2-button, rather than 
3-button, and just use Shift-click (if I remember correctly) to pick 
atoms.
If you're very used to the wheel-click to pick, you could also remap 
the wheel button to Shift-Click, and from there switching to 
2-button layout will allow you to click the scroll wheel to pick 
atoms. You won't get the benefits of the full 3-button layout, but 
it'll save you having to hold a key down every time you want to pick 
an atom.

Jacob

>Dear Warren and fellow users,
>
>I am having trouble with atom picking on windows NT, using
>pymol 0.80 (with python 2.1).
>
>If I start from the methane fragment I find it hard to pick
>the H atoms, while I can select them all I have to click
>some way away from the atom to select it (actually I find I
>need to click close to the bond between C and another H atom)
>
>BTW I have one of the MS mice which has the middle mouse replaced
>by a wheel.. mildly inconvenient for pickling with, but as far as
>I can see this is not the problem.
>
>Similar picking works on my SGI using a recent CVS checkout,
>(a month or so old, sorry I don't have an exact checkout date).
>I have noticed similar situations arising on the SGI in the past
>which trying to set up the (lb) selection, but they are not so
>consistent and I cannot reproduce them now.
>
>Any ideas?
>
>thanks in advance,
>
>Paul
>
>--
>======================================================================
>Dr Paul Sherwood,         email:  p.s...@da...
>CLRC Daresbury Lab,       tel:    +44-1925-603553
>Warrington,               fax:    +44-1925-603634
>WA4 4AD,  UK.             http:   http://www.cse.clrc.ac.uk
>======================================================================
>
>_______________________________________________
>PyMOL-users mailing list
>PyM...@li...
>https://lists.sourceforge.net/lists/listinfo/pymol-users

-- 

Jacob Corn
Gladstone Institute of Neurological Disease
415.695.3745
jc...@gl...

[PyMOL] atom picking on windows

[Paul S. <p.s...@dl...>]

Dear Warren and fellow users,

I am having trouble with atom picking on windows NT, using 
pymol 0.80 (with python 2.1).

If I start from the methane fragment I find it hard to pick
the H atoms, while I can select them all I have to click
some way away from the atom to select it (actually I find I 
need to click close to the bond between C and another H atom)

BTW I have one of the MS mice which has the middle mouse replaced
by a wheel.. mildly inconvenient for pickling with, but as far as
I can see this is not the problem. 

Similar picking works on my SGI using a recent CVS checkout,
(a month or so old, sorry I don't have an exact checkout date). 
I have noticed similar situations arising on the SGI in the past
which trying to set up the (lb) selection, but they are not so
consistent and I cannot reproduce them now.

Any ideas?

thanks in advance, 

Paul

-- 
======================================================================
Dr Paul Sherwood,         email:  p.s...@da...
CLRC Daresbury Lab,       tel:    +44-1925-603553
Warrington,               fax:    +44-1925-603634
WA4 4AD,  UK.             http:   http://www.cse.clrc.ac.uk
======================================================================