Open alex-hh opened 6 months ago
I met the same problem too. I traceback this error and noticed that character X
is not a valid token for esm. But use single-chain PDB is ok without any X
I also encountered the same problem, and I found that it is related to the complexity of the protein model. When I input a small PDB file, it can accurately predict, but once I input a large PDB file, errors will occur
Thanks for the very interesting work and the code!
I wanted to try to get the Protein MPNN CMLM model running on one of the example PDB files.
I copied the code from the Readme, but appear to have issues with tokenising the 'X' amino acids in 3uat.pdb
This is the code:
This is the traceback:
would really appreciate any help in resolving / tips for workarounds!