Kay's measure represents the spread of local minima in which refinement
can get trapped. It is a component of uncertainty, but not a whole one.
If one would use refinement that could avoid being trapped in the local
minima, then Kay's estimate would be zero!
One should also note that the B-factor is a measure of atom position
uncertainty. It is the sum (in squares) of two uncertainties: the
measurement uncertainty and the lack of reproducibility between
different unit cells. One has to remember that the B-factor unit is a
bit unusual - it is a variance (square of sigma) measure, scaled up by
8*pi*pi. One can easily reverse this to get sigma out of the B-factor.
This sigma applies to each coordinate. B-factor uncertainty should be
added to Kay's measure (in squares).
Best wishes,
Zbyszek
On 2025-11-10 14:08, Kay Diederichs wrote:
A more useful estimate, in my eyes, of coordinate precision is to
shake the coordinates after convergence of refinement, and re-refine.
Do that a couple of times and determine the rms differences of atom
positions. This will show high precision (low rms) for buried residues
with low B-value, and conversely for surface residues.
Best wishes,
Kay
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
