RE: [ccp4bb]: Refining Ions - Mn+2 Vs Mg+2...

["Savvas N. Savvides" <savvas.savvides@vub.ac.be>]

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

Raji,
there is a good possibility that your Mn2+ does not have full occupancy.
Hence the better behavior in R-free with lighter atoms (eg. Mg).  How are
the B-factors behaving with Mn2+ in the model?  I would guess unreasonably
high.  I suggest the following crystallographic experiment to estimate the
occupancy of your putative metal.  This can be done only if you have good
data, resolution better than 3 angs, and well ordered residues coordinating
to the Mn2+.

(1) refine your structure to convergence with a Mg2+ that you know behaves
well.

(2) calculate an average B-factor for the metal-coordinating residues

(3) substitute your metal with Mn2+, and reset the B factor of the metal to
20 angs-sq.

(3) try a series of short refinement runs (15 cycles of minimization +20
cycles of individual B-factor refinement) using different occupancies for
the Mn2+. Something like 0.25 0.50 0.75 1.00.  Then check to see what
B-factors this analysis gives you for the metal.  The occupancy that yields
a B-factor that is close to the average for the coordinating residues is a
good estimate of the true occupancy.

All the best
Savvas

--------------------------------------------
Savvas N. Savvides
Department of Ultrastructure - VIB
Free University of Brussels
Pleinlaan 2
Building E, room 4.4
1050 Brussels
Belgium
Tel. +32-2-629.18.49  Fax. +32-2-629.19.63
Email.  savvas.savvides@vub.ac.be
--------------------------------------------



-----Original Message-----
From: owner-ccp4bb@dl.ac.uk [mailto:owner-ccp4bb@dl.ac.uk]On Behalf Of
raji
Sent: Friday, January 31, 2003 12:27 AM
To: ccp4bb@dl.ac.uk
Subject: [ccp4bb]: Refining Ions - Mn+2 Vs Mg+2...


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

Hi,
I am trying to incorporate ions into my model.
We know from previous data that we have Mn+2 ions in our structure. Most of
us
in the lab who have tried to refine the ions as Mn+2 in CNS
(as per CNS tutorial) find that the R-factor and R-free increase (as much as
1% sometimes) and have continued to refine them as waters.
It's the same situation whether it's Mn+2 or Mn.
However, when I tried to refine all my Mn+2 as Mg+2 (by mistake),
I found that the program handles it nicely and that there is a nice drop in
R-factor and R-free.
So, I am currently refining Mn+2 as Mg+2.
Is there any explanation to what might be going on ? Does something need to
be
done differently to refine them as Mn+2 ions.
Any words of wisdom/advice/suggestions/criticisms ?
Thanks very much,
Raji

Rajeswari Edayathumangalam
Dept. of Biochemistry & Molecular Biology
Colorado State University
Fort Collins, CO 80523 USA