[phenixbb] Occupancies

Nathaniel Echols NEchols at lbl.gov
Wed May 20 10:00:40 PDT 2009

>         I am new to phenix. I am tring to refine the occupancies  
> using the phenix.refine gui. In the refinement settings- refinement  
> strategy : occupancies , when I define the groups for individual  
> atoms as chain A and resseq 1:200 (for protein atoms) and a ligand  
> group in constrained as chain B and resseq 1:10 and chain C and  
> resseq 1:10, the refinement fails with several errors. The  
> refinement runs fine when I do not touch the "occupancies" panel on  
> the gui.

Actually, right now the GUI doesn't handle all of the options for  
occupancy groups - they were too complicated to generate the interface  
automatically.  I'll see if I can fix this.

What are the errors you get, and could you please describe your  
structure in more detail?  Based on the above, I don't think you want  
to refine occupancies for chain A - you only need to do this for atoms  
that have alternate conformations or that really aren't 100% present  
(e.g. ions, also some larger ligands).  Is the ligand group in two  
conformations or do you really have two different chains?

> How do I refine in Phenix the occupancies defining as well as  
> partitioning the molecule as individual, groups and coupled groups?

Right now you'll need to do this on the command line, but if you can  
wait a day or two I may have this part of the GUI fixed.  In version  
1.4-3 and above, the parameters are this:

refinement.refine.occupancies {
	individual = None
	remove_selection = None
	constrained_group {
		selection = None

All of these may occur multiple times.  The documentation says:

individual: Selection(s) for individual atoms. None is default which  
is to refine the individual occupancies for atoms in alternative  
conformations or for atoms with partial occupancies only.

remove_selection = Occupancies of selected atoms will not be refined  
(even though they might satisfy the default criteria for occupancy  

constrained_group = Selections to define constrained occupancies. If  
only one selection is provided then one occupancy factor per selected  
atoms will be refined and it will be constrained between predefined  
max and min values.

Constrained groups are only required when you have two different sets  
of atoms (e.g. different chains, or different residues in a single  
chain, or a ligand/water and a protein chain, etc.) whose alternate  
conformations are coupled.


Nathaniel Echols
Lawrence Berkeley Lab
NEchols at lbl.gov

More information about the phenixbb mailing list