[sharp-discuss] refinement after SIR

[Clemens Vonrhein]

Hi,

if you already run autoSHARP with SIR, then the native and derivative
data will be on same origin (actually they always are, because there
is no such thing as different origins of datasets - only the indexing
could be different).

>From autoSHARP you should have a YourIdentifier-unique.data.mtz file
in your sharpfiles/datafiles directory. This contains the native and
derivative data -scaled and with same indexing.

You could refine against two datasets

  1. F/SIGF from YourIdentifier-unique.data.mtz
     HLA/HLB/HLC/HLD from eden.mtz

     This has the advantage that you are refining against 'real' data
     for which you have all the usual statistics (Rsym, I/sigI etc).

  2. FP/SIGFP from eden.mtz
     HLA/HLB/HLC/HLD from eden.mtz

     This has the advantage that the Hendrickson-Lattmann coefficients
     are consistent with the amplitude you are refineing against.

The FreeR_flag you could pick from any both of these two files. But if
you built into the map after density modification
(i.e. eden_flat_XY.Zpc.mtz) you should use the FreeR_flag from
eden.mtz probably.

 ------------------------------------------------------------

If you want to include your model in some way into the experimental
phasing and/or density modification:

  1. place your current PDB file into sharpfiles/datafiles with a name
     model_WhatEverYouWant.pdb.

  2. go back to the 'Phase Improvement ...' panel (where you did the
     density modification - or where autoSHARP did the density
     modification).

  3. run another density modification (maybe with optimizing the
     solvent content), using the model_WhatEverYouWant.pdb for the
     first N cycles (N could be 1 if you are really worried about
     bias, but I would go for 5-10). Remember to use a slightly
     different solvent content (e.g. 0.544 instead of 0.543) -
     otherwise it will complain that a run with this solvent content
     has already be done.

I hope this gives you some ideas - if not: please give a few more
details.

Cheers

Clemens

On Fri, Aug 03, 2007 at 12:05:19PM +0100, Sudharsan Sridharan wrote:
> Hi all,
> 
> I've been able to run sharp using peak dataset (3.2 A) and a native
> dataset (2.5 A). I've also built as much as I can 2 molecules of the
> asu in the resulting map. I'd like to start refining against the
> native data. How can I set the native data to the same origin as the
> map I have from Sharp to refine and do any further density
> modifications (if possible in Sharp, how ?). Is this possible in
> Sharp. or can I refine against any relevant field from the eden mtz
> file the current map is made from ? sorry, I'm new to doing MIR/MAD
> phasing !
> 
> Thanks,
> Sid.
> 
> _______________________________________________
> 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)
***************************************************************