Hi Joe,
I have a crystal structure which I believe involves the substrate and product of the enzymatic reaction occupying the same space within the active site. Ideally, I would like to refine their occupancies to sum to a value less than or equal to 1, as one would for alternate conformations of the same molecule. Phenix does not recognize that they both have partial occupancies however, and the steric repulsion consequently drives them apart during refinement. Does anyone have any advice on achieving this?
Did you consider assigning "altloc" identifiers in column 17 of your PDB file? Note that phenix supports multiple residue names for the same resid. See the newsletter article that just came out: http://www.phenix-online.org/newsletter/ Maybe all you need to do is assign the same resid to the substrate and the product, and give one altloc "A" and the other "B". Then the occupancies are automatically constrained to add up to one. Ralf