*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
First of, the average of ANY map is zero unless you include the F000
term which we never measure. The tricky part is to scale the maps so you
are comparing apples with apples. As Ian already indicated the map RMSD
is not very well defined and depends for instance on solvent content,
data completeness etc. What you want to do is scale the maps based on
their "signal" eg the density for the protein, rather then scaling the
RMSD which contains both signal and noise. One approach would be to
place an atom of the appropriate type at your position of interest and
refine just its occupancy with SHELX. This will give you the occupancy
relative to the rest of the protein atoms, all of which typically have
an occupancy of 1.0
This way you can compare different proteins, different crystal forms of
the same protein, or different data sets of the same crystal form, all
without having to worry about scaling anything.
Bart
Artem Lyubimov wrote:
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Re: RMSD of the map. Actually, I'm comparing peak values of several
different maps. To make these maps comparable, I've normalized the
maps, which sets the average density to 0 and the standard deviation to
1.0. I am now trying to establish whether the peaks at the same atomic
positions vary significantly from map to map or not.
I suppose should've given this information in my original post. Oops!
Art
==============================================================================
Bart Hazes (Assistant Professor)
Dept. of Medical Microbiology & Immunology
University of Alberta
1-15 Medical Sciences Building
Edmonton, Alberta
Canada, T6G 2H7
phone: 1-780-492-0042
fax: 1-780-492-7521
==============================================================================
