the R-factor is a complicated thing that depends on things like > weighting factors and solvent model - several things that are not part > of your model and that need to be adjusted by the refinement program > during refinement. >
I think a more accurate is "are not part of your ATOMIC model". Bulk-solvent contribution may account from 10 to 90% of unit cell content. The fact that you don't stick it into your PDB file does not mean it's not part of your model. Moreover, if you deposit Fmodel - the total model structure factors Fmodel = scale*(Fcalc_atoms + Fbulk_solvent + Fother), then they are your model representation in reciprocal space, and in fact, are more complete source of information than a bunch of points in a PDB file. Pavel
