Try turning off real space refinement:

refine.strategy=individual_sites+individual_adp+occupancies

(and at this resolution, anomalous_group refinement of the seleniums would probably also be useful, but you'd have to use the anomalous data as input.)

-Nat


On Thu, Nov 7, 2013 at 1:12 PM, Mark Wilson <mwilson13@unl.edu> wrote:
Dear PHENIX users and developers,
I am refining a structure containing SeMet residues at atomic (1.05 Ĺ)
resolution.  The model was initially refined in Refmac5 and behaved as
expected (i.e. reasonable R factors and agreement with maps).  I've used
this final Refmac5 model as a starting point for refinement in PHENIX
1.8.4 with standard settings, as below.

phenix.refine data.mtz \
refinement.input.xray_data.labels="F, SIGF" model.pdb \
 main.use_experimental_phases=false  \
 refinement.main.number_of_macro_cycles=5 output.prefix="model_refine">
phenix_refine_1.log


Strangely, the conformations of the MSE residues are changed quite
significantly after refinement in PHENIX, departing both from electron
density and from the prior model. Any help would be appreciated.
Best regards,
Mark

Mark A. Wilson
Associate Professor
Department of Biochemistry/Redox Biology Center
University of Nebraska
N118 Beadle Center
1901 Vine Street
Lincoln, NE 68588
(402) 472-3626
mwilson13@unl.edu




_______________________________________________
phenixbb mailing list
phenixbb@phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb