On 24 Oct 2012, at 11:15 AM, Nathaniel Echols wrote:

On Wed, Oct 24, 2012 at 8:07 AM, Patrick Loll wrote:

...

Curiously, when I look in the relevant directories in

...Contents/phenix-1.8.1-1168/chem_data/chemical_components

I can find cif files corresponding to ALL of these D-amino acids, including ORD (...chemical_components/o/data_ORD.cif). Files are also present for DGL, DAS, and DPN. However, these files don't contain all of the information I'm accustomed to seeing in monomer libraries (e.g., explicit bond lengths & angles), so I'm not sure if the same information is somehow encoded differently (via the idealized coordinates, perhaps?), or if I'm barking up the wrong tree.

So: Are these the actual monomer libraries that phenix.refine consults?

No. The chemical components CIFs are basic descriptions of molecules in the PDB, with accurate topology and some geometry information, but they do not include restraint targets. eLBOW can generate the restraints for you given a 3-character residue ID if it's in that directory, however. The actual built-in restraints live either in chem_data/mon_lib (the CCP4 monomer library), or chem_data/geostd (a manually curated set, some modified from the monomer library, some created entirely with eLBOW).

Sorry to be thick, but how then can I run a refinement job using an input PDB that contains DGL, DAS, and DPN, when files corresponding to these monomers are not found in either chem_data/mon_lib/d or chem_data/geostd/d? I see that these residues are listed in mon_lib_list.cif, but I can't find the actual restraints files. I'm hoping to use one or more such files to help inform my efforts to create a modified restraints file, so my questions, albeit annoying, reflect more than mere idle curiosity... Pat --------------------------------------------------------------------------------------- Patrick J. Loll, Ph. D. Professor of Biochemistry & Molecular Biology Director, Biochemistry Graduate Program Drexel University College of Medicine Room 10-102 New College Building 245 N. 15th St., Mailstop 497 Philadelphia, PA 19102-1192 USA (215) 762-7706 [email protected]

Partick You can run ReadySet! on your model PDB to get accurate D-peptide restraint CIFs. Or you can use REEL to edit the L-peptide restraints. $PHENIX//chem_data/geostd/g/data_GLN.cif etc Or you can run eLBOW with the three-letter code? If the atoms are named in a similar fashion to the L-peptides (C,N,CA,O) then the linking is automatic. Cheers Nigel On Wed, Oct 24, 2012 at 11:36 AM, Nathaniel Echols wrote:

On Wed, Oct 24, 2012 at 9:40 AM, Patrick Loll wrote:

...

Sorry to be thick, but how then can I run a refinement job using an input PDB that contains DGL, DAS, and DPN, when files corresponding to these monomers are not found in either chem_data/mon_lib/d or chem_data/geostd/d? I see that these residues are listed in mon_lib_list.cif, but I can't find the actual restraints files.

I'm hoping to use one or more such files to help inform my efforts to create a modified restraints file, so my questions, albeit annoying, reflect more than mere idle curiosity...

Amino acids (and nucleotides, and some other common monomers) live in chem_data/geostd, e.g. chem_data/geostd/g/data_GLU.cif. I think you could turn this into D-glutamine restraints fairly easily, the key step being changing the chiral center to "positiv". I'm not sure whether anything else needs to be done to get the chain connectivity working properly, however.

Normally what you would do for a residue present in the chemical components but not mon_lib or geostd is run eLBOW with the three-letter code.

-Nat _______________________________________________ phenixbb mailing list [email protected] http://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 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov

On Wed, Oct 24, 2012 at 2:31 PM, Georg Mlynek wrote:

Dear All, MR rosetta crashed (too much programs running I guess) during rosetta rebuild and it tried to reattach but somehow did not work.

I have a rescoring_results.pkl and a place_model_results.pkl in my rosetta folder.

So I want to start with rosetta rebuild again.

1. Under Job_control I changed Start Point to rosetta rebuild. 2. When adding the rescoring_results.pkl I get following error message: PHENIX did not recognize the file type for /.../.../rescoring_results.pkl

What am I doing wrong?

Are you using the GUI? If so, you're not doing anything wrong, I am - this input isn't currently supported, but I will add it ASAP. (You may have to wait another few days for a new build, however, as some other stuff is currently broken.)

Is there a workaround?

Again, assuming you're using the GUI: set up all of the parameters except the results file, then instead of clicking "Run", select "Edit parameters and run", which depending on how the GUI is configured may require looking in the Actions menu. This will pop up a text editor, and you should set the mr_rosetta_solutions parameter to the full path of rescoring_results.pkl. Then save and exit the text editor, and continue. If you're using the command line, I guess this is a question for Tom... -Nat