[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [ccp4bb]: B-factor and resolution
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html ***
(To put this discussion back in public, forgot to CC the last time...)
The main point that we have to make to non-crystallographers is that
structures are models, i.e. best educated guesses based on diffraction data.
Most Molecular Biologists will never look at B-values (or occupancies for
that matter!), there is some room for education there I think. Should we
consider some kind of graphical representation of B-Factors (Like model
emsembles as used in NMR structures)? Only crystallographers and modellers
read the REMARK sections!
From: Edward Berry [mailto:firstname.lastname@example.org]
Sent: Friday, January 05, 2001 11:06
To: Flip Hoedemaeker
Subject: Re: [ccp4bb]: B-factor and resolution
Flip Hoedemaeker wrote:
> Well, I might get myself on slippery territory here, but there's more to
> B-values I think that lowers the correlation mentioned, especially:
> - How does the crystallographer decide on the resolution cut-off? This is
> NOT universally done!
> - How are B-values refined? I think there are huge differences between the
> different refinement programs, although I never tried to refine the same
> dataset with say X-PLOR and REFMAC.
> Aren't B-factors, especially at lower resolution (say above 2.3 Ang) not
> more of a weighting scheme for correction of disorder (mosaicity), actual
> mobility, and undetected multiple conformations rather than actual
Yes, I guess I grossly oversimplified the problem in my first few lines.
But I think it questionable whether all these factors together could
completely swamp out the effect of B-factor on resolution. And your last
point, that B-factors at low resolution shouldn't be taken literally,
is actually the main point I wanted to make, and then ask "can we do
or should we acknowledge this fact in a way that the non-crystallographers
who use our structures will understand it, with REMARKS in the PDB file?