Re: dm-averaging

[Joachim Meyer <Joachim.Meyer@EMBL-Heidelberg.de>]

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

Dear Jan,

On 16.11.1999 at 19:20:36, Jan Abendroth wrote:
> Dear all,
> once again a question concerning a slightly complicated case of dm and
> ncs-averaging.
> 
> The situation:
> Within the the asymmetric unit of my crystal there are six copies of the
> protomer grouped as three dimers A, B and C. 
> I could determine the following symmetry operators:
> inter-dimer: A -> B
>              B -> C
> intra-dimer: A1 -> A2 (from subunit 1 of dimer A to subunit 2 of dimer
> 1)
>              B1 -> B2
>              C1 -> C2

IMHO, the cleanest solution would be to use a monomer mask (say, around A1),
one identity transformation and 5 further transformations:

	A1 -> A2
	A1 -> B1
	A1 -> B2
	A1 -> C1
	A1 -> C2

If your molecule consists of two domains, you would need 2 non-overlapping masks
and 2 x (1+5) rot/trans operators.

There is an intrinsic problem in using multimer masks: you assume that all
subunits within the mask are related by proper symmetry and if there were
deviations, the resulting gradients would cancel out during refinement of the 
rot/trans operators. Better use a monomer mask and analyse the operators after
refinement.

Regards,
	Joachim.

-- 
Joachim Meyer			jmeyer@embl-heidelberg.de
 	phone +49-6221-387495	FAX +49-6221-387-306
	EMBL Heidelberg, Biological Structures
	Meyerhofstr. 1, D-69117 Heidelberg, Germany
	http://www.embl-heidelberg.de/~jmeyer