[phenixbb] XYZ refinement vs individual sites refinement

[Nathaniel Echols]

On Sun, Mar 13, 2011 at 10:35 AM, Yuri <yuri.pompeu at ufl.edu> wrote:
> I am new to phenix and I am trying to probe two residues that are putative
> H-bond partners, after refinement they seem to be put in orientations that
> would make it impossible for them to be interacting with a ligand. My SA
> composite omit map may suggest that one of the Asn could be in a different
> position.
> My question: If I manually change its conformation is it wise to keep its
> position fixed during refinement to see what the new maps would look like?
> If so where would I do that under the real space or individual sites tab on
> GUI phenix? I am not sure what these 2 tasks accomplish exactly.
> Any guidance would be appreciated.

I don't know whether this is advisable - I suspect not - but for
future reference, the way to do this is:
1. Select "Individual sites" or "Real space" from the "Modify
selections for:" drop-down menu in the Refinement Settings tab.
2. Enter the atom selection you want to refine - by default this will
be blank, which means "everything", but in your case, you probably
want something like this:

not (chain A and resseq 205)

assuming that the Asn of interest is A205.

The captions here are perhaps a little confusing - "Individual sites"
means "Refine XYZ using reciprocal-space target", while "Real space"
means "Refine XYZ using real-space [map] target".  Because they modify
the same set of parameters, there is a single atom selection for both
strategies, which may not be obvious in the GUI.  I'll add a caption
to clarify this.

-Nat