[phenixbb] Combining phases?

Randy Read rjr27 at cam.ac.uk
Thu Jun 26 06:35:21 PDT 2014

Dear Yuri,

There are two basic approaches you can take.  If the MR solution is convincing then I think the best one is MR-SAD, where you use the MR solution to seed the anomalous substructure determination.  This is described in the following online Phenix documentation: http://www.phenix-online.org/documentation/reference/phaser_ep.html.  This approach has two advantages: first, the process of finding the anomalous scatterers should be more sensitive when you start with some information from the MR model than when you start from nothing.  Second, if it works you will automatically get a solution in the right hand and with the right origin.  However, if the model is too poor or incomplete then there may not be enough signal to get convincing peaks in the log-likelihood-gradient maps.

The second approach is to see whether your MR solution agrees with the map you get from SAD phasing (which would validate the correctness of both of them).  You can assess this by running phenix.get_cc_mtz_pdb, which finds the choice of origin that maximises the correlation between the model density and the density in the MTZ file.  If the choice of hand is ambiguous (i.e. because there’s not a big difference in the confidence that AutoSol assigns to the two hands), you’ll have to try this procedure with both hands of the SAD solution.  If you do get a clear answer, then you can take the MR model (offset.pdb) offset to agree with the SAD map of the appropriate hand and combine that with the HySS sites in the MR-SAD approach, and Phaser will then give you combined phases.

Good luck!

Randy Read

On 25 Jun 2014, at 20:38, Yuri <yuri.pompeu at ufl.edu> wrote:

> Dear All,
> Im struggling with a structure. I have a good native data set with resolution 2.1A and a data set with ok stats collect at low energy trying to extract some anomalous signal from intrinsic sulphur atoms (6/ASU/400aa).
> Using a model of one of the domains I got a MolRep solution that has somewhat convincing features in the electron density maps some portions of the structure but not in others. I also ran HySS and obtained some coordinates for the 6 sulphur atoms but the map here is uninterpretable. Xtriage suggests the anomalous is only good until about 8A so I am not that surprised.
> Is there any way I can try to combine my putative MolRep model and phases with the sulphur anomlaous signal and maybe get better maps to build into?
> If so, how would I do that using PHENIX?
> thanks for the help
> -- 
> Yuri Pompeu, Ph.D.
> _______________________________________________
> 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

More information about the phenixbb mailing list