Re: [ccp4bb]: Re: [o-info] R-sym I/sigI etc...

[Bart Hazes <bhazes@ualberta.ca>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

On Thu, 1 Aug 2002, Edward A. Berry wrote:

> There was a ccp4bb thread ("a simple question of resolution")
> a few months back discussing useful indicators of data resolution,
> and I think the consensus was for the "optical" resolution
> as calculated by e.g. CCP4 sfcheck. sfcheck calculates resolution
> and b-factor based on the width of the Patterson origin peak.
> This is calculated from the unphased data, making it suitable
> for a crystallization paper where you can't give R-free in the
> last shell.
> In my experience optical resolution is higher (smaller number)
> than even the most liberal d-spacing cuttoff one would use
> (after all optical resolution is measuring a different thing)
> and so might unfairly impress a referee who is not aware of
> this. Does anyone know a theoretical relation between the
> optical resolution of a dataset and the D-spacing at which
> it is truncated, supposing strong data all the way?

This sounded interesting to me so I decided to let SFCHECK loose on my
favorite dataset, 100% complete to 1.64Angstrom (only F000 is missing of
course) and 99.5% complete to 1.63 angstrom. SFCHECK estimates the optical
resolution as 1.34 Angstrom even though there is not a single reflection
beyond 1.63. Interesting...

Next I checked on the graphics if I could actually start to resolve bonded
atoms since, depending on the contour level, I can imagine you can resolve
atoms that are closer than the resolution of the data. Indeed I could resolve
some bonded atoms but this represented only a minority. Of course the optical
resolution is independent of phases whereas the ability to resolve atoms in
real space does depend on the quality of the phases. So perhaps with perfect
phases the optical resolution corresponds, at least on paper, with the minimum
atomic distance that can be resolved.

I guess the details can be read in

A.A.Vaguine, J.Richelle, S.J.Wodak.  SFCHECK: a unified set of
    procedure for evaluating the quality of macromolecular stracture-factor
    data and their agreement with atomic model.
    Acta Cryst.(1999). D55, 191-205

Bart

===============================================================================

Assistant Professor
Dept. of Medical Microbiology & Immunology
University of Alberta
1-15 Medical Sciences Building
Edmonton, Alberta, T6G 2H7, Canada
phone:	1-780-492-0042
fax:	1-780-492-7521

===============================================================================