[ccp4bb]: combining MAD data

["David E. Timm" <dtimm@iupui.edu>]

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

Hello,

1) We are trying to solve a 72 kDa structure in P21212.  We tried a Se MAD
experiment, but have only been able to identify 4 sites using a variety of
manual and automated methods.  4 Se in a 72 kDa a.u. gives an estimated
signal less than our lowest binned R-factors and this experiment hasn't
worked.

2) We also collected Se MAD data for a second crystal that is nonisomorphous
with that described above (a axes vary by > 20%), again too few sites to
solve.

3) We also collected Br MAD data using a quick dip in cryo solution
containing 0.5 M Br- and have identified 12 sites of varying occupancy.  The
unit cell here is isomorphous with that in 1).  The MAD Br data following DM
yields a partially interpretable map.  Anom. diff. Fouriers using the Br
phases and the Se DANO's confirm the Se sites in 1).

Finally, the question is what would be the best strategy for combining the 3
MAD datasets?  One possibility is to combine 1 & 3 in a single MLPHARE run
and forget about 2.  However, would it be better to use multicrystal
averaging to combine one fair and two poor sets of phase estimates (maps)?
If so, could someone send an appropriate DMMULTI script?  Other suggestions
would, of course, be welcome.

Cheers,

Dave Timm
Indiana Univ. School of Medicine
Dept of Biochemistry
Medical Sciences Building
635 Barnhill Drive
Indianapolis, IN 46202