[Chimera-users] graphics problems, BIOMT, selecting by element, etc.
meng at cgl.ucsf.edu
Wed Apr 15 09:26:26 PDT 2009
On Apr 15, 2009, at 5:30 AM, Cédric Govaerts wrote:
> Hi Elaine,
I will put responses below, after your individual questions...
> I'm not posting my question to the mailing list because most of it
> might be
> due to my machine.
> I've been encountering some problems running chimeras and I suspect a
> graphic card issue(s).
> Mostly it's the display that goes bananas, lows the machine ending
> (sometimes) in a chimera crash.
It's OK -- even if due to the machine, other people have these
problems too and might want to see the discussion.
> Here are some of the issues :
> 1/ When I zoom in/out using my mouse wheel sometimes all the
> displayed atoms
> change from wire to CPK representation. I must explicitely change it
> back to
> wire to go back.
> Normal ?
> 2/ When changing the display I have sometimes some very odd graphic
> behavior, including part of the window turing colored or black. I
> attach a
> couple of screenshots. This started just after typing a display
> command. For
> a while everything became very slow until I tried to save the image
> chimera itself which seemed to reset the graphics back to normal.
> As you can see on the screenshots, some Photoshop tool windows (PS3
> running in the backgournd) appear on the image (although not visible
> chimera itself) which leads me to think that it's a graphics card
> issue or a
> graphic card+Photoshop issue.
These definitely sound like graphics driver issues. Definitely not
normal behavior! Things to try are to update your graphics driver as
possible, and/or use the Debug Graphics Driver dialog to selectively
disable OpenGL features. One way to show this dialog is to go to
Preferences (under Favorites), General category, set "debug OpenGL on
startup" to true, quit, and restart Chimera. Documentation of the
debug dialog and discussion of graphics driver problems:
> 1/ Can I easily build a biomolecule from the PDB file ?
> I tried with
> i) MultiScale Models but could only get low res surfaces
> ii) Unit Cell but got the copies from the SMTRY records of the PDB
> file, could not specify the BIOMT
> I believe the "sym" command should work but I dunno how to make it
> read the
> BIOMT matrix in the PDB file.
For building using BIOMT:
If you use Multiscale Models, you can force loading of the atomic
coordinates for all copies by selecting All chains in the multiscale
dialog and then changing the Style to anything other than surfaces.
You can also use the "sym" command. If your structure with BIOMT
records is model 0, for example, command "sym 0 0":
> 2/ Is there a way I can select specifically all Mg atoms ? Or -
> trickier- all
> calcium (Ca) atoms ?
You can use all the end entries in the Select... Chemistry menu also
in the command line. These symbols are in the element submenu. In
cases where there is ambiguity, for example H appears both under
element and under IDATM type, you can disambiguate like
It will work to use element.Mg and element.Ca ... it's just more typing.
> 3/ When I select with the mouse, I only select one atom at a time.
> would be the easiest/fastet way to display the residue that the
> atom belongs to ?
> A/ Can I do it with the command line (does the command line
> recognize the manually selected atoms?) ?
> B/ I can use the menus and do Actions->Atoms->Backbone->Full and
> then Actions->Atoms->Sidechain->Show, but that's a bit cumbersome to
> do each
> time I wanna see a residue it seems..
If you have some atom or bond in a residue selected, just hit the
keyboard up arrow to broaden the selection to the whole residue. That
is the same as using the command:
You can specify the current selection regardless of how it was created
(with mouse, Select menu, etc.) in commands with the word "sel"
"selected" or "picked".
For example, pick one atom, hit up arrow, use command "display sel" to
display the whole residue. Actually if you use "rlabel sel" (show
residue label for any selected atoms) you don't even need the up arrow
> That's it for now.. brr, thanks !
Elaine C. Meng, Ph.D. meng at cgl.ucsf.edu
UCSF Computer Graphics Lab and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
More information about the Chimera-users