[phenixbb] Q. Phase extension

[Randy Read]

Hi,

I think a reasonable possibility would be to follow Mark's suggestion to do rigid-body refinement of the polyAla model, but only use the result from that to decide whether or not the two crystals are significantly non-isomorphous.  If the polyAla model moves during the rigid-body refinement and gives a significantly better agreement with the data (preferably measured by the likelihood score of the refinement program), then it will be worthwhile worrying about the non-isomorphism and finding a way to use the density itself as the model, not the very incomplete polyAla model.  If things don't move in the polyAla rigid-body refinement, then the standard approach of combining the HL coefficients from the derivative with the high resolution native data should be very good.

Randy

On 26 Jun 2013, at 16:24, "Terwilliger, Thomas C" <terwilliger at lanl.gov> wrote:

> Hi Nat,
> I suggest that the phase extension plus the availability of experimental phases at low resolution is key here.  The poor model will get refined (yes, not very good phases there), combined with experimental phases (still a bit weak), but then extended gradually to high resolution (possibly greatly improving the phases), and then model-building is done.  I think it could work.
> All the best,
> Tom T
> 
> On Jun 26, 2013, at 9:20 AM, Nathaniel Echols wrote:
> 
>> On Wed, Jun 26, 2013 at 2:05 AM, Mark J van Raaij
>> <mjvanraaij at cnb.csic.es> wrote:
>>> Can't you just refine the poly-Ala model against the 1.5A native data? (Perhaps after rigid body refinement or even molecular replacement if the data is too non-isomorphous). And then run Autobuild with these phases?
>> 
>> If the model only contains around 50% of residues, and no sidechains,
>> the phases from refinement are going to be fairly poor.  I've never
>> had much luck trying to autobuild starting such low completeness, even
>> with very good data.
>> 
>> -Nat
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb
> 
> _______________________________________________
> 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