This should do it (if not please get back to me off-list with sending data and model files):
Refinement settings -> Ligand linking , then Options->Link metals . phenix.refine does not trust anything in PDB file header except CRYST1 records. Pavel On Fri, Mar 6, 2015 at 10:17 AM, Kgosisejo, Oarabile <o.kgosis...@usask.ca> wrote: > Hello all, > > I am refining the structure of a metalloenzyme with 2 Mn, a cacodylate and > 5 residues coordinating the metal center (the active site) using 2.25 A > data. I am able to place the residues in their electron density using Coot. > However, after refinement using Phenix the metal-coordinating residues and > the cacodylate molecule move out of their electron densities (they pull > towards the metal center). Please see attached screenshot. The rest of the > molecule is fine > > The edits and cif files used in refinement were created using Phenix > ReadySet. The refinement options checked in Phenix are XYZ coordinates, > Real space, Individual B-factors and Occupancies. X-ray/ stereochemistry > and X-ray/ADP weights are optimized. > > I am wondering if there is refinement strategy or anything that I can do > to keep the residues in the electron density? > > Your suggestions are appreciated > > *Best Regards,* > > *Oarabile M. Kgosisejo* > *o.kgosis...@usask.ca <o.kgosis...@usask.ca>* >
