Hi Flip,

sfcheck will consider some twinning relationships not investigated by the
servers

for sfcheck version 7:

%sfcheck
outputdir/
FILE_F my.mtz
OUT U
TEST Y
<blank line>

This will generate a postscript report including a potential twin law, if
found.
As you've implied other factors can lead to an apparent relationship between
reflections, including scaling in the wrong space group, and this
possibility will have to be eliminated separately.

Also examine your spot morphology - if some of the streaky spots look
'split', then you may have a case of non-merohedral twinning.  A
particularly
frustrating protein I worked with, that I strongly suspect as being
non-merohedrally twinned,
would give a appreciably different apparent twin fraction when I tweaked
parameters at the
spot integration stage, consistently across several crystals.

Good luck!
James

Dr. James Irving
NH&MRC C.J. Martin Fellow
Division of Structural Biology
Wellcome Trust Centre for Human Genetics
Oxford University
Roosevelt Drive,
Oxford OX3 7BN
UK
email: [EMAIL PROTECTED]
phone: +44 1865 287 550

Reply via email to