[phenixbb] Refining low resolution structure

Pavel Afonine pafonine at lbl.gov
Tue Apr 29 10:44:07 PDT 2008

Hi Humberto,

I suggest that you try to use TLS in combination with group isotropic B 
(one B per residue). In my experience this almost always improves the 
model. It is important that you make a thoughtful decision about the TLS 
groups rather than refine the whole molecule as one TLS group (which is 


On 4/29/2008 10:30 AM, Humberto Pereira wrote:
> Dear all
> I was trying to refining a low resolution structure with Phenix.
> The crystal belongs to the SG P6122 with one monomer per ASU. The 
> monomer contains
> approximately 270 residues. There isn't any NCS operators.
> The crystal diffracts up to 4.0 Angstrons, with good statistics, 
> however I have only 4204 unique reflections.
> I solved the structure by MR by Phaser and the solution was quite 
> reasonably.
> Initially I performed a phenix refining cicle with SA + rigid body, 
> however the Rs values is very divergent (43 and 59%),
> so I decided to use the key 
> xray_data.r_free_flags.ignore_r_free_flags=true in order to save my 
> precious reflections with and without
> SA, and I also use the group_adp., in those cases the R is about 39-40%.
> I want a help to decide the best refinement strategies for low 
> resolution refinement with phenix.
> BTW I'm trying to produce some different constructions of the gene, 
> expecting improve the resolution.
> Thanks
> Humberto
> ------------------------------------------------------------------------
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20080429/e4eb99dc/attachment-0003.htm>

More information about the phenixbb mailing list