Hello,
I am currently trying to analyze a crystal structure that has various N-linked Glycans attached and would like to be able to work with these glycans in PyRosetta. However, if I try to work with the structure without cleaning it I receive the error message:
>>>pose = pose_from_pdb("my_molecule.pdb")
core.chemical.ResidueTypeSet: Finished initializing fa_standard residue type set. Created 9172 residue types
ERROR: unrecognized aa NAG
ERROR:: Exit from: src/core/io/pdb/file_data.cc line: 1424
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
RuntimeError: unidentifiable C++ exception
Is there currently defined residue types that I can add to my pyrosetta library that define these glycans or ways to work with uncleaned pdb files?
Any help or suggestions would be greatly appreciated.
Thanks,
Justin
Hello, Justin,
I am actually actively developing the means to add such functionality into Rosetta, but it is a work in progress. If you have a recent Linux or Mac version of PyRosetta, much of it has already been added, but the features are incomplete, untested, and unpublished.
There are 3 flags you would need to use to enable the functionality. Within PyRosetta (again, only the Linux and Mac versions), you would use rosetta.init("-include_sugars -read_pdb_link_records -enable_branching").
You also need to change the PDB file so that the string "NAG" is changed to "Glc". (Rosetta will figure out that it is a modified GlcNAc automatically.)
You also need to maintain the LINK records in the PDB file.
You also need to change a few atom names in the PDB file for the GlcNAc records to match those used by Rosetta (as can be seen in the Rosetta database in the patch file for N-acetylated sugars).
It is not entirely straightforward yet, because bugs are being worked out and various helper scripts are still being written. It is probably best if you e-mail me directly, and I can collaborate with you in getting what you want to work to work.
~Labonte <JWLabonte@jhu.edu>
Hello,
I am also extremely interested in the study of N-glycoproteins. I wonder if there has been any updates since the last comment. Also, is there a way to make it work on PyRosetta on Windows?
Thanks,
Tony
Hello, Tony,
Yes, I am fairly certain that the newest Windows PyRosetta release should now include the carbohydrate functionality I mentioned above.
I will restate again, however, that it is a work-in-progress, incomplete, untested, and unpublished. Feel free to e-mail me directly if you have any questions.
~Labonte <JWLabonte@jhu.edu>