[ccp4bb]: refinement problems

[Chloe Zubieta <chloe@perutz.salk.edu>]

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


I recently collected a MAD data set which had alot of radiation
damage (resolution started at 2.9, last data set the resolution was 3.5,
and mosaicity had also increased from 0.8 to 1.2).  I used CNS to find the
heavy atom sites, although the maps were very poor and I was not able to
resolve my spacegroup ambiguity (p43212 or p41212).  I used SHARP to
generate better looking maps and got an interpretable map (spacegroup
p43212) which had alpha helices and beta structure.  I built a
partial polyalanine model into this map and tried refining with cns using
the solvent flattened F's and Phi's from SHARP.  My attempts at refinement
were not successful and the Rfactors were extremely high, with the Rfree
never going below 53%.  Using the same data and refining in refmac gave
Rfree of 48%, but I can't seem to get any improvements.  Any suggestions?

I also took an old native data set, (different spacegroup p212121) and ran
epmr on that and got a solution which seemed reasonable.  Refining that
gave Rfree 47%, Rworking 44% so far.  I don't think the solution is
random, but I'm not sure what's wrong.

Thanks for any advice!

Chloe Zubieta
Salk Institute 
La Jolla, CA 92037