- This topic has 4 replies, 2 voices, and was last updated 11 years, 2 months ago by Anonymous.
September 18, 2012 at 1:15 pm #1409Anonymous
I apologize in advance if this has been answered somewhere else, but I’m having trouble running the backrub program with non-canonical amino acids. Our designed protein includes a chromophore and I constructed a .params file that works with the fixbb application but I can’t get backrub to recognize it. I tried renaming one of the atoms to act as a C-alpha for the pivot moves but it still doesn’t work. I was just wondering if anyone knows how to get the .params files to work with backrub or if this is even the right was to recognize non-canonical amino acids. The chromophore doesn’t even need to move at all during design, it just needs to be considered. Any help would be greatly appreciated!
September 18, 2012 at 4:44 pm #7808Anonymous
I need to know more specifically where and how it is failing. Best guess, your new residue type has garbage molecular mechanics atom types assigned, and the mm angle term used in backrub is choking. Next best guess, you need to provide three-body angle molecular mechanics parameters for any new atom types defined by your new residue (preferentially from exhaustive review of mm potentials in the literature, but just making it up from biophysical intuition or whole cloth will probably work).
September 19, 2012 at 12:59 pm #7819Anonymous
Thanks for your response. Yeah, I think you’re probably right. It crashes after reading the PDB file and outputs this error message: “core.mm.MMBondAngleLibrary: (0) No parameters for X-X-X” so it looks like it’s not finding MM angle parameters. How do I define those? Does it belong in the .params file? I’ve just been using a .params file that came directly out of molfile2params.py.
September 19, 2012 at 2:21 pm #7820Anonymous
There are five columns in the ATOM section of the top part of your params file (example is tryptophan):
ATOM N Nbb NH1 -0.47
ATOM CA CAbb CT1 0.07
ATOM C CObb C 0.51
ATOM O OCbb O -0.51
ATOM is a parsing directive. The next column is the PDB name. The next is a Rosetta atom type used for hydrogen bonding and a few other things. The fourth is the molecular mechanics (mm) atom type. I think the last is a partial charge.
You almost certainly have all Xs in your fourth column. You need to fill those in with (preferably correct) molecular mechanics atom types. The allowed atom types are in rosetta_database/chemical/mm_atom_type_sets/fa_standard. I’m pretty sure they’re CHARMM types, so I guess use CHARMM’s rules to assign them. The angles appear to be in par_all27_prot_na.prm in that directory; just the list of atoms is in mm_atom_properties.txt.
You can probably assign most of your atoms by analogy to the existing canonical and noncanonical residues in your database (rosetta_database/chemical/residue_type_sets/fa_standard/residue_types/.
If none of the atoms in your new residue are allowed to move as part of backrub, you can reassign all the mm atom types as VIRT (or just the sidechain) to dodge calculation of mm angles. This will cause problems if those atoms are directly moved by the mm-sensitive backrub step, of course.
September 20, 2012 at 1:14 pm #7827Anonymous
Yeah, that was the problem. I changed the fourth column to VIRT and it works now. I’ll try experimenting with real values later on. Thanks for your help!
- You must be logged in to reply to this topic.