On Saturday, 31 January, 2015 00:51:08 Jurgen Bosch wrote:
> Hi Ethan,
> DM with a mask of the ligand in chain A then applying the NCS matrices and
> looking at the resulting density perhaps ?
Real-space refinement is not the issue.
That can be done easily in Coot by calculating an NCS-averaged map
and then fitting the ligand into it.
I'm specifically interested in how to keep or introduce
the equivalent NCS restraint during subsequent refmac refinement.
> In Refmac you can also add a NCS restraint per residue - I assume a ligand is
> considered a residue but I have not tried this.
Yes, but that doesn't actually help.
It restrains the ligand copies to look like each other internally,
but does not restrain their binding pose relative to the surrounding
protein.
Ethan
--
Ethan A Merritt
Biomolecular Structure Center, K-428 Health Sciences Bldg
MS 357742, University of Washington, Seattle 98195-7742
