*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
The sites are amost certainly wrong, I usually try to get CC > 30% and
CC(weak) > 15% before proceding further with SAD phasing. I recommend
combining the datasets and running shelxd again reducing the resolution
cutoff in steps of 0.1A until you get a jump in CC and CC(weak) - these
will however generally increase as the resolution is truncated. A sharp
drop in occupancy after the last correct site is a very good indication.
You should also increase the number of shelxd trials (e.g. NTRY 10000) -
I have several times only had one correct solution in 10000 tries.
Another test for borderline cases, if you have NCS, is to see if you can
detect NCS for the sites from the crossword tables in the .lst file (or
using other programs that detect NCS).
George
[EMAIL PROTECTED] wrote:
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Hi: I am using Ccp4i (v 6.01) and started with a pair of SAD datasets (.sca
files from denzo) and then ran the shelxc/d/e suite with them. I got some
probably mediocre hits (cc<12%) but decided to use the best atoms from the
shelx hits and refine 'em with mlphare -- the sca file I described I ran
through scalepack,truncate and cad to get the mtz file to refine against the
shelx sites. I have made sure I'm refining anom.occupancy only (Se atoms) but
I find the sites and their occupancy are being consistently refined to 0,
often within ~8 cycles. Any ideas what I may be doing wrong?
Thanks for your help
Alex
--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-3021 or -3068
Fax. +49-551-39-2582
