[phenixbb] on refine related to b-factor

Dale Tronrud detBB at daletronrud.com
Sat Jul 25 08:32:43 PDT 2015

Hash: SHA1

   B factors are parameters which are our attempt to represent the
amount of "disorder" for that atom.  When compared to the average of
the B factors a large B factor indicates that that atom has more
motion, on either a long time scale (partial occupancy) or a short
time scale (vibration) than the rest of the protein.

   To ask if there is a refinement strategy that would lower the B
factors is the same as asking for a strategy that would move the atom
to the left.  The goal is to get the correct B factor, not the lowest
B factor.

   When you have a region with high B factors you can sometimes make
the map more understandable to the human eye by "sharpening" the map.
 This is an artificial lowering of the B factors just to make the map
easier for the user to comprehend, when they have been trained with
higher resolution maps.  This can be done inside Coot when you
calculate your map from the Fourier coefficients in your mtz file.

Dale Tronrud

On 7/25/2015 7:40 AM, Smith Liu wrote:
> Suppose a 500-aa protein crystal structure has a flexible loop of
> 30 aas, and the loop fragment has a relatively high b-factor (above
> 100) in comparison with the whole protein b-factor. Is any refine
> strategy we can refine so that the b-factor of that loop fragment
> will be decreased?
> Smith
> _______________________________________________ phenixbb mailing
> list phenixbb at phenix-online.org 
> http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe:
> phenixbb-leave at phenix-online.org
Version: GnuPG v2.0.22 (MingW32)


More information about the phenixbb mailing list