Hi Alex, thanks for the file - I will have a closer look. Did you mean the restraint did not work or it worked but you wanted something else? (I'm just trying to understand if this is a Phenix problem that we need to fix). Pavel On 10/18/12 10:30 PM, Alex Theodossis wrote:
Hi Pavel,
thanks for the info. With it I was able to correctly define the symmetry-related links as per this example:
refinement.geometry_restraints.edits { bond { action = *add atom_selection_1 = name ZN and chain C and resname ZN and resseq 1 atom_selection_2 = name NE2 and chain A and resname HIS and altid A and resseq 39 symmetry_operation = -x+1,y+1/2,-z+1/2 distance_ideal = 2.030000 sigma = 0.050 }
Unfortunately, this didn't have the desired result. Instead, the addition of riding hydrogens fixed the problem. I've also attached the relevant fragment of the refined model as requested.
Cheers,
Alex
On 17 October 2012 13:04, Pavel Afonine
mailto:[email protected]> wrote: Hi Alex,
Nigel is the best person to reply, so he may provide more comments. My understanding is that phenix.metal_coordination command (or whatever the equivalent click in the GUI is) should account for symmetry related atoms. If not you can fix it manually. First create edits file using phenix.metal_coordination and then add symmetry related bonds as described here:
http://www.phenix-online.org/documentation/refinement.htm#anch353
Also, could you please send me the PDB file or its portion containing relevant to your question atoms and I will have a closer look?
Pavel
On 10/16/12 6:33 PM, Alex Theodossis wrote:
Dear bb,
I am trying to refine a Zn ion coordinated by symmetry related residues. To avoid groups being pushed away. I have use phenix.metal_coordination to generate the necessary link edits. However only non-symmetry related interactions are accounted for. Is there a relevant option I can switch on? Alternatively, does anyone have an example of the syntax required to define the necessary symmetry operations in the elbow.edits file?
Thanks in advance.
_______________________________________________ phenixbb mailing list [email protected] mailto:[email protected] http://phenix-online.org/mailman/listinfo/phenixbb