Re: [phenixbb] pdb waters
Hi, yes, the waters in your file do not match the Monomer Library definition. If you simply replace O1 with O that will fix the problem. For example: GOOD record: ATOM 22792 O HOH X 1 37.309 34.504 -14.174 1.00 38.28 O BAD record: ATOM 22792 O1 HOH X 1 37.309 34.504 -14.174 1.00 38.28 O
I have tried running elbow on my pdb already with no luck. using your command I get the error message: elbow.builder: error: no such option: --do-all
That makes me thinking that you are using an old version of CCI Apps or PHENIX. Are you using the latest one? Is the command you run looks like: % elbow.builder x.pdb --do-all Nigel: do you have any comments? Pavel.
Thanks for all help so far..... OK, renaming the water to O from O1 worked fine, I also updated CCI Apps, now --do-all works. It only picked out one hetatm though so then I ran elbow naming the residues and it produced cif files for all my ligands - Hooray! Then I put all these files into the command line with model and structure factors and get: Sorry: Fatal problems interpreting PDB file: Number of atoms with unknown nonbonded energy type symbols: 51 Please edit the PDB file to resolve the problems and/or supply a CIF file with matching restraint definitions, along with apply_cif_modification and apply_cif_link parameter definitions if necessary (see phenix.refine documentation). Also note that elbow.builder is available to create restraint definitions for unknown ligands. Again! Even though I now have cif file for all. Any thoughts greatly appreciated. BTW refinement without ligands has provided a great map and v.good stats! Really want to refine with ligands though, cheers, Matt. On 15 Mar 2007, at 19:41, Pavel Afonine wrote:
Hi,
yes, the waters in your file do not match the Monomer Library definition. If you simply replace O1 with O that will fix the problem. For example:
GOOD record: ATOM 22792 O HOH X 1 37.309 34.504 -14.174 1.00 38.28 O
BAD record: ATOM 22792 O1 HOH X 1 37.309 34.504 -14.174 1.00 38.28 O
I have tried running elbow on my pdb already with no luck. using your command I get the error message: elbow.builder: error: no such option: --do-all
That makes me thinking that you are using an old version of CCI Apps or PHENIX. Are you using the latest one? Is the command you run looks like: % elbow.builder x.pdb --do-all
Nigel: do you have any comments?
Pavel.
Matthew Bowler MRC Dunn Human Nutrition Unit Wellcome Trust / MRC Building Hills Road Cambridge CB2 2XY Tel: 0044 (0) 1223 252826 Fax: 0044 (0) 1223 252825
Hi Matthew, Sorry for this! Sounds boring. To debug this any more I need more information, otherwise this standard error message doesn't tell me much what exactly phenix.refine doesn't like. For example, could you take your ligand molecule from your PDB file and send me so I can simulate the rest of the data and run phenix.refine myself to see exactly what's wrong? Or at least the whole .log file? Thanks! Pavel. Matthew Bowler wrote:
Thanks for all help so far.....
OK, renaming the water to O from O1 worked fine, I also updated CCI Apps, now --do-all works. It only picked out one hetatm though so then I ran elbow naming the residues and it produced cif files for all my ligands - Hooray!
Then I put all these files into the command line with model and structure factors and get:
Sorry: Fatal problems interpreting PDB file: Number of atoms with unknown nonbonded energy type symbols: 51 Please edit the PDB file to resolve the problems and/or supply a CIF file with matching restraint definitions, along with apply_cif_modification and apply_cif_link parameter definitions if necessary (see phenix.refine documentation). Also note that elbow.builder is available to create restraint definitions for unknown ligands.
Again! Even though I now have cif file for all.
Any thoughts greatly appreciated.
BTW refinement without ligands has provided a great map and v.good stats! Really want to refine with ligands though, cheers, Matt.
On 15 Mar 2007, at 19:41, Pavel Afonine wrote:
Hi,
yes, the waters in your file do not match the Monomer Library definition. If you simply replace O1 with O that will fix the problem. For example:
GOOD record: ATOM 22792 O HOH X 1 37.309 34.504 -14.174 1.00 38.28 O
BAD record: ATOM 22792 O1 HOH X 1 37.309 34.504 -14.174 1.00 38.28 O
I have tried running elbow on my pdb already with no luck. using your command I get the error message: elbow.builder: error: no such option: --do-all
That makes me thinking that you are using an old version of CCI Apps or PHENIX. Are you using the latest one? Is the command you run looks like: % elbow.builder x.pdb --do-all
Nigel: do you have any comments?
Pavel.
Matthew Bowler MRC Dunn Human Nutrition Unit Wellcome Trust / MRC Building Hills Road Cambridge CB2 2XY Tel: 0044 (0) 1223 252826 Fax: 0044 (0) 1223 252825
I have followed with interest the discussions regarding refinement of proteins with ligands. I wonder if Pavel or anybody else could comment on the refinement of proteins with covalently attached ligands. I am currently handling that in CNS as I know how to write the patches. Is this possible in phenix? Coudl it be handled by generating a "new amino acid" like say a Cys with beta mercapto ethanol attached and then get the program to link this "new amino acid" via normal peptide bonds with the rest of the protein? What about a capped N-terminus? can we make the "new amino acid" say N- acetyl-Gly and get the program to put this at the N-terminus? Any help is appreciated. Marcelo ____________________________________________ Marcelo C. Sousa, Ph.D. Department of Chemistry and Biochemistry Cristol Chemistry Building. Room 100 215 UCB University of Colorado at Boulder Boulder, CO 80309 Phone: (303) 735 4341 Fax: (303) 492 5894
Marcelo Covalently bound ligands is an active area at the moment. It is possible using the latest CCI-Apps and the interface will improve in the near future. If you have a modified amino acid in the PDB which contains the rest of the protein and you have modified the residue name (or in the near future, you have non-standard atoms in the modified residue) you can run elbow.builder protein_modified_amino.pdb --do-all which will scan the PDB file for residues it does not recognise and generate a restraints library. You can then phenix.refine data.mtz protein_modified_amino.pdb elbow.XXX.all.cif to tell the refinement about the restrains for your modified amino acid. The refinement attempts to include the modified amino acid in the backbone based on the atom names (C, N, CA, O). You should check that this occurs and let me know in the event that it fails. I'm also working on a scheme were by you can have the modified amino acids in other formats. Nigel Marcelo Carlos Sousa wrote:
I have followed with interest the discussions regarding refinement of proteins with ligands. I wonder if Pavel or anybody else could comment on the refinement of proteins with covalently attached ligands. I am currently handling that in CNS as I know how to write the patches. Is this possible in phenix? Coudl it be handled by generating a "new amino acid" like say a Cys with beta mercapto ethanol attached and then get the program to link this "new amino acid" via normal peptide bonds with the rest of the protein? What about a capped N-terminus? can we make the "new amino acid" say N- acetyl-Gly and get the program to put this at the N-terminus?
Any help is appreciated.
Marcelo
____________________________________________ Marcelo C. Sousa, Ph.D. Department of Chemistry and Biochemistry Cristol Chemistry Building. Room 100 215 UCB University of Colorado at Boulder Boulder, CO 80309 Phone: (303) 735 4341 Fax: (303) 492 5894
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Fax : 510-486-5909 Email : [email protected] Web : CCI.LBL.gov
Hi Marcelo, another option to try (different from what Nigel explained) is: 1) You use eLBOW to generate a cif file for your ligand like: % elbow.builder model.pdb --do-all You will get a file like elbow.xxx.all.cif. 2) Then you run refinement as: % phenix.refine model.pdb data.mtz elbow.xxx.all.cif params where params is the file that defines the covalent bond between your macromolecule and ligand. An example of params file is (file params contains lines below): refinement.geometry_restraints.edits { bond { action = *add atom_selection_1 = chain A and resname HEM and resid 154 and name FE atom_selection_2 = chain A and resname HIS and resid 93 and name NE2 distance_ideal = 1.8 sigma = 0.01 } For more details see "Documentation" at http://phenix-online.org/download/cci_apps/ Pavel. Marcelo Carlos Sousa wrote:
I have followed with interest the discussions regarding refinement of proteins with ligands. I wonder if Pavel or anybody else could comment on the refinement of proteins with covalently attached ligands. I am currently handling that in CNS as I know how to write the patches. Is this possible in phenix? Coudl it be handled by generating a "new amino acid" like say a Cys with beta mercapto ethanol attached and then get the program to link this "new amino acid" via normal peptide bonds with the rest of the protein? What about a capped N-terminus? can we make the "new amino acid" say N- acetyl-Gly and get the program to put this at the N-terminus?
Any help is appreciated.
Marcelo
____________________________________________ Marcelo C. Sousa, Ph.D. Department of Chemistry and Biochemistry Cristol Chemistry Building. Room 100 215 UCB University of Colorado at Boulder Boulder, CO 80309 Phone: (303) 735 4341 Fax: (303) 492 5894
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
participants (5)
-
Marcelo Carlos Sousa -
Matthew Bowler -
Nigel W. Moriarty -
Pavel Afonine -
Pavel Afonine