[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [ccp4bb]: need estimated error of occupancy



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

Hmmmmm...

There are 2 issues: first, matrix inversion gives you in the
diagonal the estimated standard uncertainties, and, very important,
the covariances in the off diagonal elements. In principle inversion
works as soon as LSQ becomes non-singular, i.e. >= one datapoint per
parameter.
Whether this is very sensible yet, is another issue, but not directly
related to resolution.

Your reviewer shows a certain neglect of the fact that B and n are
HIGHLY correlated, only very good refinements with wery large data/parameter
ratio and good low & high res data allow a distinction between a truely high
B
or just a partial occupation. Without chemical constraints (like knowing
that the total occupation is one when refining split conformations) or
other restraining knowledge (sourrounding protein's Bs in some cases) there
is practically no way to discern high B or lower n (<1) at the usual protein
resolution.

Such is even nontrivial in small molecule refinement. However, if you draw
the the atomic scattering factor vs. sin(th)/lambda graph for a
high B (say 130) vs a 0.3 occupation atom for
example, you will see that the functions are qualitatively different.
This is why, given a large enough data range, the covariation between
B and n can become small enough to make statistically significant estimates
of what is truely going on. Again, I doubt that this makes much
sense in your case.

Best, br
-----------------------------------------------------------------
Bernhard Rupp
Macromolecular Crystallography and Structural Genomics
LLNL-BBRP L448                               Phone (925) 423-3273
University of California                     Phax  (925) 424-3130
Livermore, CA 94551                          email    br@llnl.gov
URL                                 http://www-structure.llnl.gov
TB Structural Genomics Consortium  http://www.doe-mbi.ucla.edu/TB
EU Mirror         http://www.ccp14.ac.uk/ccp/web-mirrors/llnlrupp
-----------------------------------------------------------------


> -----Original Message-----
> From: owner-ccp4bb@dl.ac.uk [mailto:owner-ccp4bb@dl.ac.uk]On Behalf Of
> Susan Heffron
> Sent: Tuesday, May 21, 2002 1:58 PM
> To: ccp4bb@dl.ac.uk
> Subject: [ccp4bb]: need estimated error of occupancy
>
>
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
> Hi all.
>
> We are looking for a method to calculate the "estimated error" for the
> occupancy values of the ions in our protein structure.  The occupancy
> was refined in X-PLOR or CNS.  A reviewer of our paper is requiring an
> "estimated error" of these values.  All suggestions for how to calculate
> this are appreciated.
>
>                          Sincerely,
>                          Susan
>
> --
> ------------------------------------------------------------------
>         Susan Heffron
> Dept. of Physiology and Biophysics
> University of California, Irvine
>  Irvine, CA  92697-4560   U.S.A.
>     phone:  (949) 824-4625
>     FAX:    (949) 824-8540
> ------------------------------------------------------------------
>