Processing raw data error

[wensm77]

hi！ When I use your code to process cif files, I get an error. I checked the code, you did not pass in ‘covalents’ when you finally returned it.

  File "/home/wensm/boltz/scripts/process/mmcif.py", line 1121, in parse_mmcif
    return ParsedStructure(data=data, info=info)
TypeError: ParsedStructure.__init__() missing 1 required positional argument: 'covalents'

[wensm77]

And why is msa_id set to null? This causes the msa file to be unable to be matched during trainnig.

# rcsb.py line 136
    for i, chain in enumerate(structure.chains):
        key = f"{pdb_id}_{chain['entity_id']}"
        chain_info.append(
            ChainInfo(
                chain_id=i,
                chain_name=chain["name"],
                msa_id="",  # FIX
                mol_type=int(chain["mol_type"]),
                cluster_id=clusters.get(key, -1),
                num_residues=int(chain["res_num"]),
            )
        )

[navvye]

Running into the same issue!

[kevincheng7]

same issue here