Hi,

One way of dealing with this is to place a (symmetry) unique copy of the ligand , which may be its part.
Alternatively it's possible to disable nonbonded repulsion forces for selected atoms.

Send me model and data files if still need help with this.

Pavel

On 5/20/20 01:01, Tom Vanmarcke wrote:

Dear,

We are trying to solve a protein structure that contains a ligand (polyoxometalate). We found the ligand in the electron density but through the middle of the ligand goes a 2 fold axis. For this reason, we placed the occupendency of the ligand on 0.50. When we try to refine the structure using phenix.refine, the ligand breaks down even when adding a cif restrain file.

The cif restrain file contains information about atoms, angles and bond lengths.

Is there a way to fix this problem?

Greetings