[sharp-discuss] Simple Q: FP,SIGFP in solvent flattened .mtz

Clemens Vonrhein vonrhein at globalphasing.com
Tue Aug 16 19:45:54 CEST 2011


Hi Francis,

On Tue, Aug 16, 2011 at 11:05:57AM -0600, Francis E Reyes wrote:
> When you say 'taking into account HA substitution', are the
> contributions to the heavy atoms removed so I should not see heavy
> atoms in my maps when I refine against FPsha/SIGFPsha,HL's (solvent
> flattened)?

That depends if we're talking MAD/SAD, SIR/MIR or something else. In
general with MAD/SAD the FP/SIGFP will have the heavy-atoms. But it
can be more complex and the FP/SIGFP contain an 'average' set of HA
contributions.

You also have FH/PHIH in eden.mtz: those define the point in the
complex plane at which the FP circle is centred (which has a sigma of
SIGFP).

My rough suggestion: use FP/SIGFP as long as you're still doing major
modeling/building. At the point you start worrying about adding heavy
atoms into the model you're most likely already at the end of the
refinement (or getting there): then switch to a 'real', measured
dataset ... but ensure you take the FreeR_flag from
eden_flat_XY.Zpc.mtz or eden-unique.mtz over!

Cheers

Clemens

> 
> 
> On Aug 16, 2011, at 11:03 AM, Francis E Reyes wrote:
> 
> > Clemens,
> > 
> > Thanks for the reply.
> > 
> > 
> > When using FPsha/SIGFPsha, I imagine those are recalculated with every SHARP run? i.e. If I refine a model against a set of experimental phases, rerun sharp (calculating the phases from the updated model and combining them with the new sharp phases), FPsha/SIGFPsha will be different from experimental phases used to refine the model ?
> > 
> > 
> > 
> > Thanks!
> > F
> > 
> > On Aug 15, 2011, at 1:41 AM, Clemens Vonrhein wrote:
> >> * only FP/SIGFP (or FPsha/SIGFPsha) are in sync with the HLA-HLD:
> >> important for phase-combination during density modification or when
> >> refining including those phase probabilities.
> > 
> >> * if you did refine against these FPsha/SIGFPsha: tricky to put values
> >> into table 1 or REMARK 200. You might want to switch at the end to a
> >> 'real' dataset to refine against (for which you then also have
> >> statistics to put into REMARK 200).
> > 
> > _______________________________________________
> > sharp-discuss mailing list
> > sharp-discuss at globalphasing.com
> > https://www.globalphasing.com/mailman/listinfo/sharp-discuss
> 

-- 

***************************************************************
* Clemens Vonrhein, Ph.D.     vonrhein AT GlobalPhasing DOT com
*
*  Global Phasing Ltd.
*  Sheraton House, Castle Park 
*  Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group      (http://www.globalphasing.com)
***************************************************************


More information about the sharp-discuss mailing list