[phenixbb] low res structures and Phenix

Pavel Afonine PAfonine at lbl.gov
Mon Aug 20 18:30:11 PDT 2007

Hi John,

> 1.  Can Phenix do B-factor sharpening for improved maps? 

This is not implemented in phenix.refine. CNS can do this.

> 2.  How can one optimize Ksol and Rprobe (bulk solvent model) in Phenix?

This is highly optimized in phenix.refine. I believe we have the most 
sophisticated protocol for doing this (Acta Cryst. (2005). D61, 850-855, 
"A robust bulk-solvent correction and anisotropic scaling procedure" P. 
V. Afonine, R. W. Grosse-Kunstleve and P. D. Adams).

What you can try to do is to optimize the mask calculation parameters: 
solvent_radius and shrink_truncation_radius (the default values are 1.11 
and 0.9). To do specify in the command line (or in the parameter file, 
whatever you use) one extra parameter: optimize_mask=True; for example:

% phenix.refine model.pdb data.mtz optimize_mask=True

This will tun much longer but may result in slightly improved statistics.

> 3.  Are there any other suggestions for dealing with low resolution 
> structures in Phenix?

How you refine the ADPs? TLS might be an option to try out.


More information about the phenixbb mailing list