Thanks for your quick reply! However now I get the following error message,

"Fatal problems interpreting model file: NUmber of atoms with unkown nonbonded energy type symbols: 3 Please edit the model file to resolve the problems nd/org supply a CIF file with matching restraint definitions, along with apply_cif_modification and apply_cif_link parameter definitions if necessary. Alternatively, to continue despite this problem use: stop_for_unknown=False"

the log file says,

  Monomer Library directory:

    "c:\phenix-1.8.4-1496\phenix-1.8.4-1496_sources\chem_data\mon_lib"

  Total number of atoms: 11298

  Number of models: 1

  apply_cif_link:

    data_link: NAG-ASN

      mod_id_1: DEL-O1

      mod_id_2: DEL-HD22

    residue_selection_1: chain A and resname NAG and resid 401

    residue_selection_2: chain A and resname ASN and resid 74

  apply_cif_link:

    data_link: BETA1-4

      mod_id_1: DEL-HO4

      mod_id_2: DEL-O1

    residue_selection_1: chain A and resname NAG and resid 401

    residue_selection_2: chain A and resname NAG and resid 402

  apply_cif_link:

    data_link: NAG-ASN

      mod_id_1: DEL-O1

      mod_id_2: DEL-HD22

    residue_selection_1: chain B and resname NAG and resid 401

    residue_selection_2: chain B and resname ASN and resid 74

  apply_cif_link:

    data_link: BETA1-4

      mod_id_1: DEL-HO4

      mod_id_2: DEL-O1

    residue_selection_1: chain B and resname NAG and resid 401

    residue_selection_2: chain B and resname NAG and resid 402

  Model: ""

    Number of chains: 4

    Chain: "A"

      Number of atoms: 5075

      Number of conformers: 2

      Conformer: "A"

        Number of residues, atoms: 319, 4675

          Unusual residues: {'NAG%DEL-O1%DEL-HO4': 1, 'NAG%DEL-O1': 1}

          Unexpected atoms: {'NAG%DEL-O1%DEL-HO4,HO4': 1}

          Classifications: {'undetermined': 2, 'peptide': 317}

          Modifications used: {'DEL-HD22': 1, 'DEL-O1': 2, 'DEL-HO4': 1, 'NH3': 1}

          Link IDs: {'PTRANS': 27, None: 2, 'TRANS': 287, 'PCIS': 2}

            Not linked:

              pdbres="ALA A 317 " segid="A   "

              pdbres="NAG A 401 " segid="A   "

            Not linked:

              pdbres="NAG A 401 " segid="A   "

              pdbres="NAG A 402 " segid="A   "

      Conformer: "B"

        Number of residues, atoms: 319, 4675

          Unusual residues: {'NAG%DEL-O1%DEL-HO4': 1, 'NAG%DEL-O1': 1}

          Unexpected atoms: {'NAG%DEL-O1%DEL-HO4,HO4': 1}

          Classifications: {'undetermined': 2, 'peptide': 317}

          Modifications used: {'DEL-HD22': 1, 'DEL-O1': 2, 'DEL-HO4': 1, 'NH3': 1}

          Link IDs: {'PTRANS': 27, None: 2, 'TRANS': 287, 'PCIS': 2}

            Not linked:

              pdbres="ALA A 317 " segid="A   "

              pdbres="NAG A 401 " segid="A   "

            Not linked:

              pdbres="NAG A 401 " segid="A   "

              pdbres="NAG A 402 " segid="A   "

          bond proxies already assigned to first conformer: 4309

    Chain: "B"

[...]

    Chain: "C"

[...]

  Residues with excluded nonbonded symmetry interactions: 26

    residue:

[...]

  Number of atoms with unknown nonbonded energy type symbols: 3

    "ATOM   5046  HO4 NAG A 401 .*.A    H  "

    "ATOM   6215 HD22 ASN B  74 .*.B    H  "

    "ATOM  10057  HO4 NAG B 401 .*.B    H  "

  Time building chain proxies: 4.05, per 1000 atoms: 0.36

Deleting these atoms (NAG HO4 and ASN HD22) manually doesn't seem to do the trick - as I explicitly want to refine hydrogens I reckon they are added back on. Also it seems BETA1-4 wants to desperately link NAG O4 to NAG O1 which doesn't exist .. Shuld I manually define my NAG-NAG link instead of using the BETA1-4?

Thanks again!

B

2014-06-12 2:12 GMT+02:00 Nathaniel Echols <nechols@lbl.gov>:

On Wed, Jun 11, 2014 at 5:05 PM, Benjamin Stauch <bs448c@gmail.com> wrote:

I want to refine a model that has two NAGs linked to an Asn - the sugars are already part of my initial coordinate file, but upon refinement the link between the NAGs gets lost. I have defined a cif_link file as follows that I load (using the GUI),

     apply_cif_link {

       data_link = "NAG-ASN"

       residue_selection_1 = "chain A and resname NAG and resid 401"

       residue_selection_2 = "chain A and resname ASN and resid 74"

     }

     apply_cif_link {

       data_link = "BETA1-4"

       residue_selection_1 = "chain A and resname NAG and resid 401"

       residue_selection_2 = "chain A and resname NAG and resid 402"

     }

Okay, I think this may be part of the problem - you are missing the outer "scope" for these parameter blocks, so they are not processed correctly.  If you change "apply_cif_link" to "refinement.pdb_interpretation.apply_cif_link" in both places it may fix the problem.

Refinement doesn't crash or produce errors as far as I can see, however when checking the MolProbity output after refinement, it complains of missing atoms O1 in NAG 401 and 402. (Not sure if this is relevant)

Sorry, this is actually my bug (those atoms should not be included in covalently linked sugars), and I think I still have an email complaining about this in my inbox from late 2012.  I need to overhaul that bit of code anyway so I will see if Nigel and I can come up with something smarter.

-Nat