Protein names ending with numbers produces error in load_confidence_data in colabfold_process_output.py

[nt314p]

Hi,

Because the load_confidence_data function uses regex to parse parameters from pdb file names, the regex has the possibility to match with the protein name derived from the fasta file. If this matching occurs, the parse is incorrect or fails.

Would it be possible to pass data associated with the pdb files in a file instead of the name? A csv file could be used.

Thanks!