Hi Emmanuel,
try MZ protocol (MZ = multi-zone) in phenix.refine. It does the smart
rigid body refinement starting from a few low resolution reflections and
ending up with using all data. In-between it does the maximum-likelihood
bulk-solvent modeling and scaling. All together it results in a VERY BIG
convergence radius. So, if the problem is really not the mis-assigned
unit cell or similar, then you have a good case for MZ protocol to try
out. The whole protocol is fully automated, so you'll get the "yes/no"
answer in one run.
For more information goto: http://www.phenix-online.org/download/cci_apps/
and look: 1) "QuickFacts" to get an overview; 2) "Documentation" (link
right next to phenix.refine) to get started.
Let me know if you have any questions.
Good luck,
Pavel.
Emmanuel Prata wrote:
Dear all,
I have a data set of a protein soaked with sacarose, and the statistic
data is good until the molecular replacement (Rmerge 0,07 (24%),
I/Sigma 7 (2,7), completness (99,8 (98%)), and resolution 1,98A. After
this step, I can not get the Rfactor (64%) to drop (I believe the
spacegroup is correct).
Could anyone give me suggestions?
Thank you in advance,
Emmanuel Prata
