Hi Hailiang,

Sure! Maybe I can prepare Fc with solvent modeling and anisotropic scaling
first. However, I am not sure how to generate the U_overall matrix in your
formula. The U_overall given by the pdb header is acturally in Cartesian
coordinate, and a "diagonalization matrix" is needed to do a
transformation before applying to h/ht. Not sure how to generate the
"diagonalization matrix" (still doing research:-(, any help will be

can't we just use a PDB file or it is not available for whatever reason in your context?
Like I said before, anisotropic scaling can be done without model, so you don't have to do it. If you still want to, then everything is described in this paper:

P.V. Afonine, R.W. Grosse-Kunstleve & P.D. Adams. Acta Cryst. (2005). D61, 850-855. "A robust bulk-solvent correction and anisotropic scaling procedure"

Good luck!
Pavel.