Dear Brian,
Thank you for your reply. I think I could have provided you with the
correct formula myself ;-) but it is indeed correct now. In any case, the
verbal description I gave in my message specified it already.
The definition in the mmCIF dictionary does look like the source of
that aberrant formula. Is it actually used somewhere as defined? That would
be scary.
In any case, the one formula in the paper by Jones et al. cited in the
mmCIF description is a real-space R-factor, not a real-space correlation
coefficient. Some sort of monstrous recombination seems to have taken place
between the formula for an R-factor and that for a correlation coefficient,
giving rise to the eye-popping formula I spotted.
I hope you won't mind if I post this reply to the CCP4BB again, just
for its general entertainment value. It isn't every day that one comes
across this type of typesetting hybridisation :-))) .
With best wishes,
Gerard.
--
On Tue, Mar 10, 2020 at 05:39:54PM +0000, Brian McMahon wrote:
> Dear Gerard
>
> I've made an amendment following a version of the equation sent
> separately by Randy Read. Please review and let me know if that is
> satisfactory.
>
> This error seems to have arisen from a transcription of a definition in the
> mmCIF/PDBx dictionary
> (http://mmcif.pdb.org/dictionaries/mmcif_pdbx_v50.dic/Items/_struct_mon_prot.RSCC_all.html)
> which will also need updating. I'll follow that up.
>
> Thanks for bringing this to our attention.
>
> Best wishes
> Brian
>
>
> On 10/03/2020 17:11, Gerard Bricogne wrote:
> > Dear Brian,
> >
> > Sorry to have taken this matter up in such a visible manner: I noticed
> > this very wrong formula in someone's paper, and that person then told me
> > where he had found it. Having landed on that page, I didn't know where to go
> > next. I can't create an account because I do not have an entry in the World
> > Directory of Crystallographers.
> >
> > What is the best way of getting this formula corrected?
> >
> >
> > With best wishes,
> >
> > Gerard.
> >
> > ----- Forwarded message from Gerard Bricogne <g...@globalphasing.com> -----
> >
> > Date: Tue, 10 Mar 2020 14:15:33 +0000
> > From: Gerard Bricogne <g...@globalphasing.com>
> > Subject: [ccp4bb] An error in the IUCr Online Dictionary of Crystallography
> > To: CCP4BB@JISCMAIL.AC.UK
> >
> > Dear all,
> >
> > While we are all discussing how to "push the frontiers" of our field,
> > there seem to be some dark and dusty corners in the documentation of some
> > basic notions. For instance the formula for the real-space correlation
> > coefficient given at
> >
> > https://dictionary.iucr.org/Real-space_correlation_coefficient
> >
> > is completely garbled. The absolute value signs in the denominator are
> > superfluous, although harmless since we are dealing with real numbers, but
> > the numerator is wrong in two respects:
> >
> > 1. there should not be absolute values around the deviations from the
> > mean of the two types of densities;
> >
> > 2. the expression should be a single summation of products of those
> > two
> > types of deviation, not a product of single sums!
> >
> > Does anyone know how to get this corrected?
> >
> >
> > With best wishes,
> >
> > Gerard.
> >
> > ########################################################################
> >
> > To unsubscribe from the CCP4BB list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
> >
> > ----- End forwarded message -----
> >
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
