Re: [ccp4bb] how to refine RNA with two phosphate at 5' by phenix

[Ralf W. Grosse-Kunstleve]

In the most recent phenix.refine version, the links should be automatic if the
backbone atom names are consistent with the naming for standard RNA residues.
Let me know if you're having problems with phenix 1.6 or 1.6.1.
Ralf



----- Original Message ----
From: William G. Scott <wgsc...@chemistry.ucsc.edu>
To: CCP4BB@JISCMAIL.AC.UK
Sent: Fri, April 2, 2010 5:07:09 PM
Subject: Re: [ccp4bb] how to refine RNA with two phosphate at 5' by phenix

Lisa Wang wrote:
> hi all,
> I have one structure contains dsRNA with two phosphate group at 5'.. I
> named
> the first residue GDP and got a cif file by running phenix.elbow. Then I
> refine my structure by running phenix.refine my.pdb my.cif. But the fist
> and
> the second residue do not link with each other, that is no bond between
> O3*
> of the fist residue and P of the second residue.
> Please give me some idea. Thanks
>

You have to explicitly link them.  Something like the following:

  geometry_restraints.edits {
    bond {
      action = *add delete change
      atom_selection_1 = "chain A and resid 1 and name O3\\*"
      atom_selection_2 = "chain A and resid 2 and name P"
      symmetry_operation = None
      distance_ideal = 1.59
      sigma = 0.01
    }
    angle {
      action = *add delete change
      atom_selection_1 = "chain A and resid 1 and name C3\\*"
      atom_selection_2 = "chain A and resid 1 and name O3\\*"
      atom_selection_3 = "chain A and resid 2 and name P"
      angle_ideal = 120
      sigma = 1
    }
  }