*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Hi Phil,
Although I have only seen this once as a pathological case, it can
mean that the resolution cutoff calculated in Scalepack can be
significantly at odds with what Scalepack2mtz thinks your maximum
resolution is (using the post-refined parameters) so it pays to scale
to at least a little beyond your intended maximum resolution for
refinement or phasing.
I am not clear about the above explanation.
Say, I use I/sig, R-merge etc. etc. criteria to determine the highest
resolution. Now, I do not wish to open Pandora's box for resolution
determination. I then re-scale my data in Scalepack to that *highest*
resolution. That means my output .sca file will only contain the
reflections within the range that I specify.
As I've just discovered and if I understand correctly, there's an
inherent danger in this range being incorrectly specified since the
post-refined values of the unit cell do not get used when they should
be.
But, how would rescaling with an extended higher resolution help? I
will see the statistics deteriorate at that highest bin and still
truncate my data based on the statistical criteria. What would alert me
as to how high to extend the resolution to compensate for the above?
Could you pl. clarify.
Thanks.
Raji
Raji Edayathumangalam
Postdoctoral Associate
The Rockefeller University
Box 224. 1230 York Avenue
New York, NY 10021
