looks like your heavy atoms belong to a crystallographic symmetry-related molecule (or several). The program used to find the heavy atom sites does not necessarily make them coincide with where the automatic building program finally builds the crystallographic model.
Use the Draw => Symmetry commands in coot to make them coincide.
Or, which is a very useful geometric exercise, let some of the symmetry operators belonging to your spacegroup (plus whole cell translations if necessary) loose on the heavy atom coordinates until you can make them coincide.

On 2 Dec 2013, at 20:56, Egor Tchesnokov wrote:

Hello,
 
Level: beginner
 
1.       Data collected at wavelength of 0.9537
2.       Ran Xtriage: anomalous data seem to extend to 2.2 A
3.       Ran Phaser-MR to get partial model (1 copy in AU)
4.       Ran AutoSol using partial model and anomalous data
 
AutoSol finds heavy atom sites (FOM=0.69; Overall score=75.6+/-3.3).
Coot displays model/density without heavy atoms
Heavy atoms are displayed outside the model (using zoom out).
 
Question: what did I do wrong?
 
Thank you for your time and help.
 
Egor
<AutoSolHeavy atom sitesoutside the model.docx>_______________________________________________
phenixbb mailing list
[email protected]
http://phenix-online.org/mailman/listinfo/phenixbb