Re: ncs restraints

[Garib Murshudov <garib@yorvic.york.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***

Dr. David E. Timm wrote:
> 
> ***  For details on how to be removed from this list visit the  ***
> ***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***
> 
> Hello,
> 
> Some questions regarding refinement:  Is it advisable to use ncs 
> restraints in the refinement of a dimeric protein (2 fold NCS) phased 
> at 1.9 A using MAD data?  

I would try 2 refinement protocols. With NCS and without it. With 
positional parameters I think it should not harm but for restraints
on B-values there might be problems as 2 copies might have different
overall B-values

> Is it necessary to account for the SeMet
> anomalous signal (32 Se/92 kDa A.U.) in refinement?  (The Bijvoet
> differences were over 9% for the peak data). 
Current refinement programs (which use Maximum likelihood residual)  can
not refine aganst F+ and F-. For this shape of likelihood should be 
changed. But you could use phases derived from MAD for refinement. It
should improve behaviour of refinement. And after refinement you will
have map with combined phases.

>  Are the SeMet geometric
> restraints significantly different from Met in REFMAC (or XPLOR)? 
> Which dataset/wavelength is most appropriate for use in refinement? 

Yes they are different. 

> Is it necessary to maintain a consistent test set (for Rfree) between
> the 1.9 A data collected from the SeMet substituted crystal and a 1.5
> A native dataset?  If so, how would one accomplish this in the
> overlapping resolution bins?  

Yes it is important to use same free reflections for related crystal
structure refinement. One way of doing it:

Use cad to combine SeMet and native data sets. 
Use freerflag to generate list of free reflections. At this stage 
you should use

complete FREE=FreeR_flag

keyword to let program to know that it should complete free R list but
not generate new list


Regards
Garib

------------------------------------------------------------------
Garib N. Murshudov, Chemistry Department, University of York, U.K.
Tel: Home +44 (1904) 43 35 89, work:  +44 (1904) 43 25 65
------------------------------------------------------------------