[sharp-discuss] Can autosharp use NCS automatically?

[Clemens Vonrhein]

Dear Nian,

On Sun, Nov 18, 2012 at 07:32:05PM -0500, Nian Huang wrote:
> Although I have used the autosharp many times before, I haven't really pay
> attention to the NCS averaging. I noticed I can access the NCS searching in
> the phase improvement panel after solvent flattening.

Yes - that is using GETAX for detecting C_n/D_n symmetries. See

  http://www.ncbi.nlm.nih.gov/pubmed/10089413
  http://www.ccp4.ac.uk/html/getax.html

You can also use CCP4i for running this. The GETAX-results might need
visual inspection (of those CC-maps) to understand the correct NCS
operators to use.

There are other ways of detecting your NCS operators. One of the most
popular ones is based on the heavy atom sites. For that these must
obviously follow the NCS, so Se-MET or S (Cys, MET) are prime
candidates - but other intrinsic heavy atoms (Ca, Zn, Fe etc) can be
helpful too. When dealing with HA soaks it can become quite messy
though.

> However, can I ask autosharp to include NCS averaging before the
> run?

Yes(ish): the latest version of autoSHARP with it's new building
module (called "Long John Silver" - because it is using K. Cowtan's
parrot an buccaneeer programs) will let parrot decide on NCS
itself. This should work well if you have enough HA sites that obey
the NCS. So you get the NCS averaging as part of the automatic model
building there.

However, since automating NCS is a very non-trivial task, we would
always recommend running additional NCS averaging jobs outside of
SHARP/autoSHARP - especially in cases where your structure doesn't
just 'pop out' ...

You can start from your eden_flat_XY.Zpc.mtz file (see
http://www.globalphasing.com/sharp/manual/appendix1.html#edenflat),
your NCS operators and masks etc.
 
> The other question is about the density modification. After a search of an
> old archive about solomon vs DM in CCP4
> http://wserv1.dl.ac.uk/list-archive-public/ccp4bb/2003-10/msg00439.html
> Does that mean I can only use sharp with solomon and mlphare with DM?

You can use any averaging/density-modification program you want.

Just make sure you use the correct items from the SHARP/autoSHARP
eden*mtz files (amplitude/sigma, starting map-coefficients, HLs for
phase combination). This is very important: you need to understand
what the density-modification program is doing (especially if it does
phase combination) and how to tell it with what (currently best) map
to start.

E.g. for running DM starting from the eden_flat_XY.Zpc.mtz file you
should be using

  LABI FP=FPsha SIGFP=SIGFPsha -
       PHIO=PHIBshasol FOMO=FOMshasol -
       HLA=HLA HLB=HLB HLC=HLC HLD=HLD -
       FDM=FBshasol PHIDM=PHIBshasol
  LABO FDM=FDM PHIDM=PHIDM FOMDM=FOMDM -
       HLADM=HLADM HLBDM=HLBDM HLCDM=HLCDM HLDDM=HLDDM

and then look at maps in coot with the FDM/PHIDM columns.

Does that help? If not please let us know ...

Cheers

Clemens

-- 

***************************************************************
* 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)
***************************************************************