*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Against - probably no way to measure f' of the non-anomalous atoms (unless
there's some way to de-convolute the fluroescence scan, which is dominated
by the anomalous atoms), so you'd have to use calculated values; f' is not
resolution-dependent, so errors here can be absorbed by the scale factor
(however, this is somewhat of a fudge factor, seeing as it's unlikely that
the variation in f' for all atom types is uniform).
The "non-anomalous" contribution is the atomic scattering factor, f. f'
and f" are interdependent, (like ORD and CD are one amnifestation of the same
physical phenomenon). So you need to have a significant absorption edge
somewhere in the neighborhood in order to get an anomalous dispersion
contribution (f') that makes it look like there aer fewer electrons
present.
You can see this for your favorite absorbing atom here:
http://skuld.bmsc.washington.edu/scatter/AS_form.html
For Mn++, at about 1.4 A the maps look fine. Collected at the inflection
point, where the abs value of f' blows up, the Mn++ density is weaker than
that of the associated nitrogens and oxygens (at least by eyeball
integration).
So in practice you really have to be sitting on the inflection point for
this to be a huge problem. (The negative peak is really sharp).
Bill
