[Chimera-users] B-Factor Difference Plot

Vladimir Arinkin v.arinkin at fz-juelich.de
Thu Aug 31 01:41:51 PDT 2017

Dear Amanda,

    you also need to keep in mind one thing when comparing B-factors from different structures. B-factors refined in the structure does not only contain true Atomic Displacement Factors (ADP-factors) but also distortions of the crystal lattice, which are being modeled into the B-factors as one cannot really separate them.  Probably for this reason you assume that one structure is more "flexible", it maybe just due to higher crystal lattice distortions. You can try to normalize to mean B-factor of all protein residues in the structure.

    And the second, make sure that both structures has isotropic B-factors (no ANISOU records in PDB file) otherwise you cannot simply compare numbers.

Vladimir Arinkin

Am 29.08.2017 um 00:34 schrieb Amanda Evans Constantinides:
Hi Elaine,
Thanks for your detailed reply. I successfully generated Chimera output files containing lists of average B factor values, aligned the residues, plotted residue vs. B factor difference, and saw the graphs existed mainly on one side or the other of the x axis, but were not distributed across the x axis as expected based on what I saw using B putty. I tried also using Baverage in CCP4 and made Main Chain B factor difference plots and saw the same thing. My question, hopefully written more clearly, is how to normalize between the structures in comparison. I am looking at the equation for b-factor to see how I can accomplish this. If you have any ideas or know anyone who does, please send them my way!

Best regards,

Amanda Constantinides
Graduate Research Assistant
Georgia State University
From: Elaine Meng <meng at cgl.ucsf.edu><mailto:meng at cgl.ucsf.edu>
Sent: Monday, August 28, 2017 5:37:36 PM
To: Amanda Evans Constantinides
Cc: chimera-users at cgl.ucsf.edu<mailto:chimera-users at cgl.ucsf.edu>
Subject: Re: [Chimera-users] B-Factor Difference Plot

Hi Amanda,
Visually comparing the putty models (worms in Chimera) between the two structures is not meaningful unless you are sure you mapped the same value to the same radius in both structures.  Even so, it is hard to see in this way.

The more direct way is to just look at the PDB file in a text editor and see what the B-factor values are in the corresponding residues.  I don’t know how you are calculating the differences, but you would have to make sure your script or whatever is mathematically correct and that it is also correctly pairing the residues between the two structures.  I.e. the residue numbering might be different in the two structures.

Now, comparing all the atomic values is probably too hard, so you could use Chimera to output the average values per residue so that you have fewer numbers to compare.  In the Render by Attribute dialog (the same one used to make the worms display) you could choose File… Save Attributes and then choose to save the attribute of “residues” named “average->bfactor” to a file.  You would need to do this for each of the two structures.  So that it doesn’t include other stuff besides the protein,  you could use command “select protein” and then choose the option to “Restrict save to current selection” (or alternatively, delete all the stuff you don’t care about for this comparison like solvent before saving the values).

Now, if you are sure your script is mathematically correct and also that the correct residues are paired between the two structures, then maybe your question is really how to compare B-factors between two structures when they seem to be on a different scale or systematically offset.  The question of whether or how best to normalize the values between different crystal structures is not a Chimera question per se, but a more general biophysical one to which I do not know the answer.

Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco

> On Aug 28, 2017, at 9:00 AM, Amanda Evans Constantinides <aconstantinides1 at student.gsu.edu><mailto:aconstantinides1 at student.gsu.edu> wrote:
> Hello,
> I am plotting the B-factor differences between two similar structures. I do not understand why the plot is almost entirely on the positive side of the X-axis. If I open B factor putty models in PYMOL and in Chimera, there are some visible areas that should translate to negative B-factor differences on a quantitative plot. There is no way one structure is entirely more flexible than the other. Can you help explain this to me? Is there away to adjust the zero point to more accurately reflect these differences on the plot?
> Thank you,
> Amanda Constantinides
> Graduate Research Assistant
> Georgia State University

Chimera-users mailing list: Chimera-users at cgl.ucsf.edu<mailto:Chimera-users at cgl.ucsf.edu>
Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users

Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20170831/22519591/attachment.html>

More information about the Chimera-users mailing list