On Thu, Mar 31, 2011 at 10:14 AM, Jacob Keller < j-kell...@fsm.northwestern.edu> wrote:
> Also, > setting occupancy = zero is not fudging but rather respectfully > declining to comment based on lack of data. I think it is exactly the > same as omitting residues one cannot see in the density. No, it's not the same. If you have placed any atoms, even with zero occupancy, you have said something about where you expect the atoms to be, or at least where the refinement program thinks they should be. "Declining to comment" would be deleting them, not guessing. I think a reasonable number could be derived and agreed upon, and > would not be surprised if there is such a derivation or analysis in > the literature answering the question: > > "At what b-factor does modelling an atom become insignificant with > respect to explaining/predicting/fitting the data?" > > That point would be the b-factor/occupancy cutoff. > Although atoms with very high B-factors may have almost no impact on F(calc), if the occupancy is non-zero they will still be driven by gradients with respect to X-ray data, and their positions (or changes thereof) will in turn affect other atoms, through geometry restraints if not F(calc). So there is no point at which these atoms cease to be relevant to the task of fitting. -Nat
