[phenixbb] PHASER-TFZ scores with ensembles higher?

[Randy Read]

Dear Jan,

If you send the logfile (preferably off-list to avoid cluttering mailboxes) I can have a look to see whether it seems convincing.  Having a Z-score greater than 8 is usually a very good sign, particularly with recent versions of Phaser that account for the statistical effects of translational NCS (before which it was possible to get a very high Z-score for a spurious solution).  Some other things we like to see are that no potential solutions with comparable or higher Z-scores were rejected by the packing tests, and that the refined LLG after placing the first molecule is greater than 60.

There are certainly cases where there's an unambiguous solution but the phases are too poor to get anywhere with refinement.  This is a really active area of research.  For a good fraction of such cases, phenix.mr_rosetta increases the convergence radius enough that you can get over the refinement barrier (method and examples in DiMaio et al, 2001: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3365536/).  There's also Tom Terwilliger's morphing procedure (phenix.morph_model), and using the right refinement protocol can also help tremendously, such as the DEN refinement from Axel Brunger's group (also implemented in phenix) and jelly-body refinement implemented by Garib Murshudov in Refmac.

If that's not enough, then you'll need additional phase information, either from experimental phasing or, if you have multiple crystal forms of the same molecule, multi-crystal averaging.

The short answer is that you can solve the molecular replacement problem without actually solving the structure.  However, if the LLG is higher (which almost always goes together with a higher Z-score) the phases will also be better, but they may still not be good enough.

Regards,

Randy Read

On 28 Jun 2013, at 08:02, Jan Gebauer <jan.gebauer at uni-koeln.de> wrote:

> Dear Phaser users,
> 
> I recently tried to solve two difficult MR jobs by using ensembles in PHASER.
> With ENSEMBLES in this case I mean aligned structures prepared for PHASER by Phenix.ENSEMBLER.
> In both case I got solution with rather high TFZ-score (> 8), which I couldn't get wit individual structure, but was not able to refine them.
> 
> My question is, whether using ensembles would increase the Z-Score without really improving phases. In other words, if there is another cut-off for the "Phaser solved it" Z-score, when using ensemble in contrast to single structures...
> 
> Regards,
> Jan
> -- 
> Dr. Jan Gebauer
> AG Prof. Baumann
> Institut für Biochemie / Uni-Köln
> Otto-Fischer-Str. 12-14 / 50674 Köln
> Fon: +49 (221) 470 3212	
> Fax: +49 (221) 470 5066
> 
> http://px.uni-koeln.de/
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb

------
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical Research      Tel: + 44 1223 336500
Wellcome Trust/MRC Building                   Fax: + 44 1223 336827
Hills Road                                    E-mail: rjr27 at cam.ac.uk
Cambridge CB2 0XY, U.K.                       www-structmed.cimr.cam.ac.uk