Issue when handling PTM-residues

[dhusmann]

It looks like the script doesn't work when running on predictions that have modified residues. For example, running on a prediction that contains several Aceytyl-Lysine (three letter code "ALY") residues gives error:

Traceback (most recent call last):
File "/home/groups/ogozani/programs/AlphaBridge/define_interfaces.py", line 150, in
main()
File "/home/groups/ogozani/programs/AlphaBridge/define_interfaces.py", line 143, in main
interface_df_per_token, interface_df = define_interfaces(in_dir, mode, contact_threshold)
File "/home/groups/ogozani/programs/AlphaBridge/define_interfaces.py", line 80, in define_interfaces
list_sequence_info, rec_sequence_list, structure_sequence_list, polymer_chain_dict = FEATURE_OBJECT.extract_sequence_info()
File "/home/groups/ogozani/programs/AlphaBridge/src/module/confidance_contact_matrix.py", line 215, in extract_sequence_info
structure_sequence_list = structure.get_sequence_list()
File "/home/groups/ogozani/programs/AlphaBridge/src/module/parsers.py", line 262, in get_sequence_list
seq_dict[asym_id] += upper_protein_letters_3to1[mon_id]
KeyError: 'ALY'

[dhusmann]

would be ideal to include expanded protein letter alphabet to the one found here:
https://github.com/pnnl/PTMPSI/blob/fa78b8b8b22ad8f2caae1386b0fe428007753bf2/ptmpsi-awsem/required_packages/biopython/Bio/Data/SCOPData.py#L199

[drlemmus]

Support for PTMs is in the works but not available yet. We have a curated mapping for non-standard residues which partially overlaps with the BioPythin data.

[Dan-salv]

Sorry for the delay in my response! I’m currently working on addressing this issue. You can expect an update with the added support for ligands and PTMs by the end of this month.