Re: [ccp4bb]: origin definition

["Ralf W. Grosse-Kunstleve" <rwgk@yahoo.com>]

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

--- Bernard Santarsiero <bds@uic.edu> wrote:
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> I have a set of phases from an MIR run in SOLVE and a potential molecular
> replacment solution of about half the structure in space group P3.  They
> have different origins in Z since it's either defined by the set of heavy
> atom positions in SOLVE, or the set of coords in the MR search model.  I
> want to compare the SOLVE phased map and the MR sfall/2Fo-Fc map.  I see
> at least two possibilities:
> 1.  mask a map region around the model of the MR solution and move it
> around the phased map; the highest CC should indicate the best overlap of
> the two maps. I think I can use RAVE to do this.
> 2.  calculate a series of phases and list the 00l's with phases. Each set
> is from a set of coords that have been translated in Z. At some point,
> these phases should match those from the SOLVE solution. 

A more straightforward alternative would be to compute an isomorphous
difference map with MR phases. The peaks in this map should give you compatible
heavy atom sites. To align these with your SOLVE solution you could use this
web server:

http://cci.lbl.gov/cctbx/emma.html

Use the SOLVE solution as the "Reference model", and the peaks from the
isomorphous difference/MR phases map as the "Other model". To be safe I'd use
more peaks from the difference map than there are heavy atoms in the SOLVE
model. The web server will tell you which peaks match the atoms in the SOLVE
model. You will also get the rotation matrix and translation vector that leads
to the matches. The search is exhaustive so you might get more than one
solution. The solutions are sorted by the number of sites matched.
If you don't get solutions increase the "Match tolerance" (I'd probably change
it to 3 A from the start).

Ralf


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com