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

NCS auto-mask

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

Fred. Vellieux <vellieux@ibs.fr> wrote:

>So, for proper NCS (e.g. 2-fold) one can give an e.d. map centered on
>particle on which averaging is to be carried out and "do it" without a
>mask. The density for both zones 1 and 2, where the NCS applies, will
>improve by the averaging operation, and the density outside will

>For improper symmetry, with a map centered as above, no mask: if one uses
>the operator relating (zone 1 to zone 2) then the density for zone 1 will
>improve whereas all the rest, including zone 2, will deteriorate. And the
>opposite will take place if one uses the operator relating (zone 2 to
>zone 1).

>The same considerations also apply for the calculation of real space
>correlation coefficients.

I don't see any theoretical reason why a clever programmer couldn't
calculate CCs and automatically generate masks for improper symmetry, so
long as he/she specified a point in the monomer rather than in the
multimer, and so long as he/she maintained separate masks for each
monomer.  the point specified would only have to be inside the monomer
region, not at its center, since the bounds of the masked region would
be defined by a contiguous region with CC of averaging above some
cutoff value.  i'm sure it would be more complicated than for the case of
proper symmetry, with all sorts of things to keep in mind like overlap,
but i do not believe it would be impossible.  likely situations would
arise where the auto-masking procedure would fail, but that happens for
proper symmetry as well.

     Crystals are pretty, but they just kind of sit there,
         not doing much of anything.  - Larry Wall
                        David J. Schuller
                        modern man in a post-modern world
                        University of California-Irvine