Dear all,

On Tue, Nov 18, 2025 at 06:55:15PM +0000, Martin Moche wrote:
> Are there software or online servers that given space group and cell
> parameters and number of your measured reflections calculate the
> "efficient resolution" assuming 100% completeness in each shell?

autoPROC [1] comes with a tool called "aP_operational_resolution" that
does a slight variation of this (see also pointers by Mitch): it
doesn't just use every observation to then compute that "effective"
resolution, but selects e.g. those with an individual I/sig(I)>=2.0
[2].

The idea is to provide a single number for "significant" data. After
all, we can always convince an integration and scaling+merging program
to output numbers to whatever reciprocal space limits we decide:
having a number associated with a HKL doesn't make this "data".

There might be much better ways of judging the information content of
a merged reflection (and we also prefer looking at likelihood values
of reflections during BUSTER refinement), but ...

> Perhaps a better estimate of resolution in a single number than
> today's practice as:

... as a single number we find our approach quite useful.
 
> 1. Spherical integration, XDS and DIALS, integrating reflections
>    that are not present
> 2. Staraniso integration, having a few percent of reflections in the
>    highest resolution shell, making statements of resolution in a
>    single number impossible

Indeed.

It is difficult to describe all vegatables in your garden using as the
only measure the concept of "radius": works for peas and oranges,
pears become tricky ... and beans or cucumbers feel really
neglected. Much better to use "weight" or "volume" if one is really
forced to use a single number for everything.

Unfortunately, we still seem to insist in various reports that
everything in our garden grows ball-shaped ... I'm glad it doesn't in
mine ;-)

Cheers

Clemens

[1] https://www.globalphasing.com/autoproc/
[2] https://www.globalphasing.com/autoproc/wiki/index.cgi?OperationalResolution


On Tue, Nov 18, 2025 at 01:37:36PM -0600, Mitchell Miller wrote:
> Hi Martin,
> 
> Re: Question 2:
>   Are you wanting to calculate something similar to what Gerard Kleywegt
> called "effective resolution"
> and that he attributes the term and calculation method to a 1995 personal
> communication from Bart Hazes
> in a couple of his papers on good refinement practices and validation from
> 1997-2000.
> https://www.ebi.ac.uk/pdbe/docs/EMBO/Validation_XrayStructures_Gerard.pdf -
> 2000 - Acta D56, 249-265
> https://hod.greeley.org/papers/ByAuthor/Kleywegt/gmrp.html - preprint of
> his 1997 Methods & Enzymology
> article?
> 
> Bart Hazes' sftools program can calculate this.  See the "complete" command
> were it is called the
> "100% criterion"
> https://www.ccp4.ac.uk/html/sftools.html#checking-data-completeness
> (You could also get this 100%  resolution for different I/sigI cutoffs by
> first selecting only those
> reflections above the desired threshold before calling the complete command
> -)
> 
> Regards,
> Mitch

########################################################################

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/

Reply via email to