[Chimera-users] Rendering TLS information in Chimera ?
goddard at sonic.net
Tue Apr 24 12:32:14 PDT 2012
Display of TLS fitting of anisotropic B-factors looks like a
difficult problem. Here is a good try at it, a program called TLSView
(part of a package called mmLib), developed by Jay Painter. Chimera
does not do this.
My understanding is that TLS does a fitting of anisotropic b-factors
for a group of nearby atoms. It is based on the idea that the motions
of these atoms are correlated. It is a quadratic approximation using
position from an arbitrarily chosen origin for the group of atoms. The
fitting is predicting the anisotropic B-factor for each atom (a tensor
with 6 independent components which determine the 3 principle axes
directions and amplitudes along those directions) and it involves 20
independent parameters. Two very basic questions given TLS fitting are
1) how much does the TLS fit differ from the individual atom aniostropic
B-factors that produced the fit? and 2) what coherent motion of the
atoms would give rise to these TLS parameters? A visualization to
answer the first question would be pretty easy, but it requires having
both TLS and the individual atom anisotropic B-factors. The 1SS8 PDB
only reports TLS. The answer for the second question is harder to
visualize and TLSViewer shows 3 orthogonal screw axes that need not go
through a common origin. I get the concept but am not sure how valuable
it is. It would take some examples with specific molecular structures
and specific large scale motions interpreted via TLS for me to
understand this TLSViewer visualization better.
> Hi Steve,
> While Chimera currently understands PDB ANISOU records and can show that information with thermal ellipsoids, it does not understand TLS information in the header. I do see the TLS info in the REMARK 3 lines in pdb1ss8.ent, but I didn't see anything in the PDB format specifications.
> However.... at least from these deposition instructions, it sounds like depositors are supposed to convert the TLS information into ANISOU records beforehand. Also, the page has a server to do the conversion:
> Perhaps (at least for now) you could try using the server above, and then Thermal Ellipsoids (in menu under Tools... Structure Analysis) or command "aniso":
> I did this for 1ss8. It can be quite busy since it is per-atom, but I tried showing only CA atoms and their ellipsoids for chain A, 90% probability level shown in the image attached below.
> I hope this helps,
> Elaine C. Meng, Ph.D.
> UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
> On Apr 24, 2012, at 6:02 AM, Steven Ludtke wrote:
>> Hi. A question about rendering B-factor information: Over the last decade, rather than simple anisotropic B-factors, it has become increasingly common to represent local "motion" using the translation-libration-screw (TLS) model, which provides for anisotropic representation of such motions. Is there a good way to represent this data in Chimera ? Here is an example. If you render the PDB structure 1OEL as atoms, and color the atoms by B-factor, you can immediately observe regions of high variability in the apical domain. This same data was later re-refined using a TLS model as PDB structure 1SS8. When you render this structure in Chimera in the same way, you see high B-factors only for a few specific sidechains. The overall large-scale variability is gone, presumably because the TLS parameters have absorbed it, and the "B-factor" is now only a residual B-factor. However, I can't see any obvious way of rendering the TLS information. Any tips ?
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Chimera-users