[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Masking

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

> I think you still need a mask, because non-crystallographic symmetry
> is "local" symmetry.

In my discussion I mentioned providing a map centered on the particle.
Obviously you need some knowledge of where the particle is situated.

> the cell will no longer have the true symmetry, and calculated phases
> will give bad density even for the n-mer that obeyed the ncs.

In such a case, there is no unit cell reconstruction and no phase
calculation by inverse FFT.

You are simply interested in what the density looks like (or what the
correlation coefficients are) in the area you suppose your particle is

I stand with what I said earlier.

> Now I realize you might be talking about a single cycle of density averaging
> with no mask, not calculating phases but using the improved density
> directly to make a mask to use in the next cycle. That sounds reasonable
> to me, what happened to the sym-related multimers wouldn't matter.

In fact, you will be using the mask in the INITIAL cycle.

There is no single cycle of density averaging since the steps of
reconstructing the unit cell from the averaged asymmetric unit,
map inversion, computation of weights and recombination with the |Fo|
are simply not there!

> auto-masking may be a convenience for simple cases of NCS

I know of cases where such approaches were used for improper NCS.
Automation is not used there because with improper NCS, every case
is "special". That just meant more work for the crystallographer.

Sorry about all this lengthy discussion, but it is right that things
are presented correctly on a discussion list.

(Hi JJ by the way)


s-mail: F.M.D. Vellieux (B.Sc., Ph.D.)
        Institut de Biologie Structurale J.-P. Ebel CEA CNRS
        41 rue Jules Horowitz
        38027 Grenoble Cedex 01
Tel:    (+33) (0) 476889605
Fax:    (+33) (0) 476885494
e-mail: vellieux@ibs.fr
   Don't even think of asking me: "I would          Het leven is genieten
    if I could, but I can't" (L.R., 1998)           (to live is to enjoy)
Public key: FMDV