Dear all,

Apologies if this has been covered already in a previous post — but I can't find a suitable response in the archive.

We have the following situation...

We have soaked a ligand into an apo-crystal, collected diffraction data, and solved the structure.

What is apparent from the electron density,  is that the ligand isn't in the crystal at full occupancy, it's roughly about 70-80%.  

However, as the ligand binds, it causes a small conformation change in the active site of the protein, altering the position of around 4 amino acids.

What I want to do, and can't quite get phenix.refine to do is the following...

Refine the occupancy of the ligand (chain C resname LIG) with the "A" conformation of the active site residues (which should be the same, as they are mutually dependent) — and then refine the occupancy of the "B" conformation of the active site residues —  which should all theoretically add up to a total of 1.

Could anyone help me with how the occupancies / constrained_group parameters should be set up in this case?

With thanks,

Tony.