[ccp4bb]: Summary: water rings, ice rings

[Mark van Raaij <m.vanraaij@chem.leidenuniv.nl>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Fred Vellieux tells me XDS (latest version) has a (nice) option of
excluding resolution
bins and says he always decides himself what to exclude and never lets a
"black box" tool do it for him. 

Yu Wai Chen and Phil Evans remind me it is mosflm in which you can and
always could exclude resolution bins and scala could never do this (eg.
RESOLUTION 15.0 1.5 EXCLUDE 3.79 3.63 EXCLUDE 2.29 2.22 EXCLUDE 1.92
1.90). Guess my memory was wrong here.

They and Laurent Maveyraud advocate removing the data from the
ice-rings, while Remy Loris, Steve Prince and Flip Hoedemaeker suggest
is is better to keep the information - so I guess here opinions are
divided.

Valerie Biou writes her maps with the data from the ice-rings looked
better than using data with ice-rings removed.

>From discussion in Leiden with Jan Pieter Abrahams and Raj Pannu:
Perhaps for refinement the data without ice rings is the best, because
the refinement programs will not include missing reflections in the
target. For map calculation, the dataset with water rings may be best,
because a bad estimate for a reflection is better than setting it to
zero.

So I have integrated the data both ways (obviously the statistics
without the rings are better) and will try refining and map calculation
with both datasets and compare the results. By the way, the data is
quite redundant, overall multiplicity 6.0, so really bad outliers should
be taken care of.

Thanks to all for your suggestions and I hope I have not misquoted any
of you,

-- 
Mark van Raaij, m.vanraaij@chem.leidenuniv.nl