[phenixbb] Refinement

[Edward A. Berry]

On 03/31/2015 02:12 PM, Pavel Afonine wrote:
~~~~~~~~~~~~~~~~
>> In fact, the C atom is covalently linked to the nearest Cys, so the
>> HEM side chain should be 'moved' into the density and the green blob
>> shouldn't be there.
>
> Is this link defined anywhere in restraints used in refinement?
>
> Pavel

Here is an example of restraints for heme c.
Values are from the high-resolution structure 1C75.
Here residue 501 is the heme of course; and 37, 40, 41, and 160 are
the C,C,H, and M in CxxCH---M heme c binding motif.
I use HEC.cif for the heme parameters, but make sure the bond lengths
in the vinyl groups are consistent with single bonds.

   geometry_restraints.edits {
  
     bond {
       action = *add delete change
       atom_selection_1 = chain D and resid 501 and name FE
       atom_selection_2 = chain D and resid 41 and name NE2
       symmetry_operation = None
       distance_ideal = 1.99
       sigma = 0.1
       slack = None
     }
     bond {
       action = *add delete change
       atom_selection_1 = chain D and resid 501 and name FE
       atom_selection_2 = chain D and resid 160 and name SD
       symmetry_operation = None
       distance_ideal = 2.33
       sigma = 0.1
       slack = None
     }
     bond {
       action = *add delete change
       atom_selection_1 = chain D and resname CYS and resid 37 and name SG
       atom_selection_2 = chain D and resname HEC and resid 501 and name CAB
       symmetry_operation = None
       distance_ideal = 1.815
       sigma = 0.1
       slack = None
     }
     bond {
       action = *add delete change
       atom_selection_1 = chain D and resname CYS and resid 40 and name SG
       atom_selection_2 = chain D and resname HEC and resid 501 and name CAC
       symmetry_operation = None
       distance_ideal = 1.815
       sigma = 0.1
       slack = None
     }



>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org
>