[Chimera-users] Thermal ellipsoids are not updated when playing multi-frame PDB trajectory

Lukasz Walewski Lukasz.Walewski at ruhr-uni-bochum.de
Fri Apr 1 09:45:03 PDT 2011

On 03/31/2011 09:48 PM, Eric Pettersen wrote:
> On Mar 31, 2011, at 6:59 AM, Lukasz Walewski wrote:
>> Hi,
>> I am looking at the trajectory with thermal anisotropy factors that
>> change from frame to frame, i.e. I have a multi-frame PDB file that
>> contains ATOM and ANISOU records. The components of the anisotropy
>> tensors change during the simulation (the data I am analyzing has a
>> redefined meaning). I am able to properly read and play the trajectory.
>> Ellipsoid representation of the thermal U factors works excellent for a
>> single frame. However it seems that the ellipsoids are not updated when
>> I play the trajectory, i.e. the atoms move as expected but the
>> ellipsoids retain both their positions and shape from the first frame.
>> Is there any way around ?
> Hi Lukasz,
> Chimera doesn't store any info on a per-frame basis right now, so the
> shortest answer is no.  However, depending on the size of the system you
> are working with and the length of your trajectory (and your patience!)
> there may be a workaround.  Basically the workaround is to read your
> trajectory file using the regular FIle..Open dialog rather than with the
> trajectory tool.  This will cause each MODEL in your file to be opened
> as a separate model in Chimera.  This consumes a lot more memory than
> the trajectory would, so I would say it is practical if we are talking
> dozens of models or less, or hundreds of models if your system is no
> more than 1000 atoms or so, or a few thousand models if we're talking
> about a small molecule.
> Of course you lose the convenience of the MD Movie's playback interface.
>  Nonetheless, you can get a poor man's playback with a script like this:
> aniso parameters-you-want
> ~modeldisp
> modeldisp #0.1; some-timing-command
> ~modeldisp #0.1; modeldisp #0.2; some-timing-command
> ~modeldisp #0.2; modeldisp #0.3; some-timing-command
> etc.
> Hopefully you or someone you know is familiar with programming, which
> would allow you to create such a script easily rather than by insanely
> tedious hand editing.  You run the script either with File...Open or the
> "open" command.  The script file name should end in ".cmd" or ".com". 
> A couple of points about the script:
> 1) Commands on the same line separated by semi-colons are executed
> without the screen being redrawn between commands.
> 2) There are a couple of choices for "some-timing-command":
> a) "wait N" where N is an integer.  In that case Chimera will wait that
> many thirtieths of a second before going on to the next command.  So
> "wait 15" would pause half a second before going on to the next command.
>  Keep in mind that as such a script runs it can be paused by pressing
> Shift-Esc (and then resumed the same way) or aborted by pressing Esc.
> b) "pause" which will suspend script execution until any key other than
> Esc or End is pressed.  Pressing End will abort a "paused" script.
> I will be opening a feature-request ticket in our bug database for
> storing some information on a per-frame basis and getting Thermal
> Ellipsoids to use it, with you on the recipient list so that you'll be
> notified when it gets worked on.
>> In any case I find Chimera wonderful software that I came upon only
>> recently. Thanks guys for developing it!
> Thanks!  We're working on our NIH grant renewal right now.  Hopefully
> that will translate into several more years of funding.  Keep your
> fingers crossed!
> --Eric

Dear Eric,

Thanks for your advice and the ticket. At the moment I can live with
single frame view, since I only need a few representative snapshots. It
would be however very useful to have a dynamic thing as well, so I am
looking forward to the follow up from the feature request list.

And I definitely keep my fingers crossed for the grant renewal!

Best regards,

More information about the Chimera-users mailing list