*** For details on how to be removed from this list visit the *** *** CCP4 home page http://www.ccp4.ac.uk ***
The answer of this question lies in how accurate the anomalous signal can be measured and ,not to be underestimated, the type of anomalous scatterer one has as the value of f0-f' is important as well. f0 brings your Sim contribution. For 1 fully occupied Hg in lysozyme (and perfect data) and using the Ramachandran and Raman selection rule for which SAD phase to choose, you will be right in about 70% of the cases. This success rate goes down when fheavy^2/ftotal^2 goes down. Some addtional thoughts on this topic can be found in a recent paper, where I did some numerical integration of likelihood functions used in ML sad phasing to get <FOM> for different phasing scenarios (Acta Cryst D61, 1437-1448). Also, an old posting by Ed Berry is usefull. Follow this link: http://people.cryst.bbk.ac.uk/~ubcg09j/ccp4arc/bb2000/msg00658.html HTH Peter
