heavy atom in pdb for refinement
Hi, I have Zn atoms in my pdb for refinement. However, it failed with the following scripts. Do I have to create a libary of this heavy atom? Number of atoms with unknown nonbonded energy type symbols: 6 "ATOM 1 Zn Zn Z 1 .*. Zn " "ATOM 2 Zn Zn Z 2 .*. Zn " "ATOM 3 Zn Zn Z 3 .*. Zn " "ATOM 4 Zn Zn Z 4 .*. Zn " "ATOM 5 Zn Zn Z 5 .*. Zn " "ATOM 6 Zn Zn Z 6 .*. Zn " Time building chain proxies: 1.78, per 1000 atoms: 0.45 Here is the format of my pdb: ..... ATOM 1 Zn Zn Z 1 32.028 47.603 13.955 1.00 30.00 Zn ATOM 2 Zn Zn Z 2 29.017 71.434 8.164 1.00 30.00 Zn ATOM 3 Zn Zn Z 3 39.223 49.330 -6.810 1.00 30.00 Zn ATOM 4 Zn Zn Z 4 24.811 89.144 -4.406 1.00 30.00 Zn ATOM 5 Zn Zn Z 5 23.652 72.429 6.125 1.00 30.00 Zn ATOM 6 Zn Zn Z 6 30.132 63.054 -31.918 1.00 30.00 Zn TER 7 Zn 6 ATOM 10 CL CL L 1 20.125 63.405 10.719 1.00 30.00 CL TER 11 CL 1
Hi,
I have Zn atoms in my pdb for refinement. However, it failed with the following scripts. Do I have to create a libary of this heavy atom?
no, you shouldn't. Zn should be recognized automatically if it is properly defined in input PDB file. I just verified that if it is formatted like this: HETATM 2306 ZN ZN D 1 24.598 75.945 -1.390 1.00 5.00 ZN then it works fine. I'm not sure if capitalization is important as well as HETATM instead of ATOM - this is something you can try in no time. Pavel.
On Sep 11, 2009, at 3:54 PM, Pavel Afonine wrote:
I have Zn atoms in my pdb for refinement. However, it failed with the following scripts. Do I have to create a libary of this heavy atom?
no, you shouldn't. Zn should be recognized automatically if it is properly defined in input PDB file. I just verified that if it is formatted like this:
HETATM 2306 ZN ZN D 1 24.598 75.945 -1.390 1.00 5.00 ZN
then it works fine. I'm not sure if capitalization is important as well as HETATM instead of ATOM - this is something you can try in no time.
I'm pretty sure capitalization is indeed important, because I had a similar problem in both phenix.refine and eLBOW with manganese (MN is okay, Mn is not). ------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]
Yes, the capitalization matters. Ralf is notified. Pavel. On 9/11/09 4:08 PM, Nathaniel Echols wrote:
On Sep 11, 2009, at 3:54 PM, Pavel Afonine wrote:
I have Zn atoms in my pdb for refinement. However, it failed with the following scripts. Do I have to create a libary of this heavy atom?
no, you shouldn't. Zn should be recognized automatically if it is properly defined in input PDB file. I just verified that if it is formatted like this:
HETATM 2306 ZN ZN D 1 24.598 75.945 -1.390 1.00 5.00 ZN
then it works fine. I'm not sure if capitalization is important as well as HETATM instead of ATOM - this is something you can try in no time.
I'm pretty sure capitalization is indeed important, because I had a similar problem in both phenix.refine and eLBOW with manganese (MN is okay, Mn is not).
------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
Actually, after I change Zn to ZN, phenix.elbow does not work with the
following error message:
phenix.elbow overall_best_090909e_001-coot-0.pdb --all-residues --do-all
------------------------------------------------------------------------------
electronic Ligand Builder & Optimisation Workbench (eLBOW) 1.4 3
- Nigel W. Moriarty ([email protected])
------------------------------------------------------------------------------
Random number seed: 1653885000
0:00 Parsing Parsing Parsing Parsing Parsing Parsing Parsing Parsing
Parsing P
No unknown residues
[qwan@phaser2 AutoBuild_run_10_]$ phenix.elbow
overall_best_090909e_001-coot-0.pdb
------------------------------------------------------------------------------
electronic Ligand Builder & Optimisation Workbench (eLBOW) 1.4 3
- Nigel W. Moriarty ([email protected])
------------------------------------------------------------------------------
Random number seed: 1654559024
0:00 Parsing Parsing Parsing Parsing Parsing Parsing Parsing Parsing
Parsing P
No molecule read
Use --all-residues to view residues if this is a PDB file
On Fri, Sep 11, 2009 at 7:19 PM, Pavel Afonine
Yes, the capitalization matters. Ralf is notified. Pavel.
On 9/11/09 4:08 PM, Nathaniel Echols wrote:
On Sep 11, 2009, at 3:54 PM, Pavel Afonine wrote:
I have Zn atoms in my pdb for refinement. However, it failed with the following scripts. Do I have to create a libary of this heavy atom?
no, you shouldn't. Zn should be recognized automatically if it is properly defined in input PDB file. I just verified that if it is formatted like this:
HETATM 2306 ZN ZN D 1 24.598 75.945 -1.390 1.00 5.00 ZN
then it works fine. I'm not sure if capitalization is important as well as HETATM instead of ATOM - this is something you can try in no time.
I'm pretty sure capitalization is indeed important, because I had a similar problem in both phenix.refine and eLBOW with manganese (MN is okay, Mn is not).
------------------- Nathaniel Echols Lawrence Berkeley Lab [email protected]
_______________________________________________ phenixbb mailing [email protected]http://www.phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
Hi, if you modify your input PDB file according to the example I sent in my previous email, then phenix.refine should work just fine. In this case you should not use elbow at all.
Actually, after I change Zn to ZN, phenix.elbow does not work with the following error message:
Because now elbow "thinks" there is nothing unknown in your input PDB file. Pavel.
Hi, Why do we need both files .eff and .def? Can we edit the .eff file and use it the next round of refinement? Maia
Dear Maia,
I think this is documented...
http://www.phenix-online.org/documentation/refinement.htm#anch8
Briefly, when you run phenix.refine "the first time" e.g. like this
phenix.refine mymodel.pdb mydata.mtz
the parameters used in this refinement is written to an eff
(effective) parameter file. In this case this would be something like
mymodel_refine_001.eff
Furthermore a def (definition) file for the next run is generated.
Now, if you wanted to include water refinement you could
phenix.refine mymodel_refine_002.def ordered_solvent=true
and thus generating mymodel_refine_002.eff which will include the
information you provided on the commandline.
The short answer:
Yes, you can just edit the .eff file and use that in refinement.
Best regards,
Folmer Fredslund
2009/9/20 Maia Cherney
Hi, Why do we need both files .eff and .def? Can we edit the .eff file and use it the next round of refinement?
Maia _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
participants (5)
-
crystallogrphy -
Folmer Fredslund -
Maia Cherney -
Nathaniel Echols -
Pavel Afonine