[phenixbb] explicit solvent in low resolution refinement

[Guenter Fritz]

Dear all,

we have determined the structure of a protein complex at 3.5 A 
resolution as well as the structures of three subunits of this complex 
independently at resolutions better than 2 A. I had used the models of 
the subunits as reference models in refinement, perfect.
Now I was wondering whether one could use explicit solvent positions 
(e.g. inner shell water molecules) from the high resolution structures 
in the low resolution refinement.
This should be superior to a simple bulk solvent model.
I placed the inner shell water molecules from the high res structures 
into the low res model and added harmonic restraints to the water 
positions in low res refinement. I see a decrease of about 0.5% in Rwork 
and Rfree  versus the model without water molecules.

However, adding harmonic restraints excludes the option to use the 
reference models in refinement.

Is it possible to use protein and water atoms from the reference models 
to generate restraints for the low resolution refinement?

Thank you,
Guenter