problem with multimer mode, ValueError: Could not parse description:

[wh277]

running in multimer mode...
Finished running alignment for 1_1
Finished running alignment for 1_2
Finished running alignment for 1_3
Finished running alignment for 1_4
Traceback (most recent call last):
File "inference.py", line 548, in
main(args)
File "inference.py", line 164, in main
inference_multimer_model(args)
File "inference.py", line 281, in inference_multimer_model
feature_dict = data_processor.process_fasta(
File "/home/fy/FastFold-main/fastfold/data/data_pipeline.py", line 1165, in process_fasta
chain_features = self._process_single_chain(
File "/home/fy/FastFold-main/fastfold/data/data_pipeline.py", line 1114, in _process_single_chain
chain_features = self._monomer_data_pipeline.process_fasta(
File "/home/fy/FastFold-main/fastfold/data/data_pipeline.py", line 936, in process_fasta
hits = self._parse_template_hits(
File "/home/fy/FastFold-main/fastfold/data/data_pipeline.py", line 884, in _parse_template_hits
hits = parsers.parse_hmmsearch_sto(
File "/home/fy/FastFold-main/fastfold/data/parsers.py", line 656, in parse_hmmsearch_sto
template_hits = parse_hmmsearch_a3m(
File "/home/fy/FastFold-main/fastfold/data/parsers.py", line 627, in parse_hmmsearch_a3m
metadata = _parse_hmmsearch_description(hit_description)
File "/home/fy/FastFold-main/fastfold/data/parsers.py", line 589, in _parse_hmmsearch_description
raise ValueError(f'Could not parse description: "{description}".')
ValueError: Could not parse description: "0000|3jqh_A/8-65 [subseq from] mol:protein length:167 C-type lectin domain family 4 member M".

[Gy-Lu]

Hi, I think drop "0000|" in your description of fasta file may solve your problem, i.e.
>3jqh_A/8-65
SOMESEQUENCE

[wh277]

thanks，I will try it

[Gy-Lu]

Hello, there should be a > before the discription, I updated my comment because GitHub explained it as a quote symbol :-\

[wh277]

OK, thanks

[wh277]

It works. But another problem appears. raise ValueError(f"hit.name did not start with PDBID_chain: {hit.name}")
ValueError: hit.name did not start with PDBID_chain: 6qq_A.
6qq is a small molecule in PDB, not a protein, there are other small molecules in hmm_output.sto, not only one.
do you know how to solve it?

[Gy-Lu]

Hi, I am not sure either.
Could you share the content of your fasta file? It would be really helpful to me to debug.

[wh277]

OK, it contains 2 same sequences.

1_1
MNDQARSVLTTVLEEGAWSVRQWPGDAAAETVHSALGRVTDLGDRVQGIAYTTSGAMSVHTATPELTTRLDGRDQPIPLDSVYELRLWAVIDGRADDGLLAHELRWLNGSGTAEIRVGGQELLSANGDAGWQTTRCWTRSNSYLQHGATAPFDEPSKAMTSVEVFTEEPTYGNTVFVDEIMTGRWS
1_2
MNDQARSVLTTVLEEGAWSVRQWPGDAAAETVHSALGRVTDLGDRVQGIAYTTSGAMSVHTATPELTTRLDGRDQPIPLDSVYELRLWAVIDGRADDGLLAHELRWLNGSGTAEIRVGGQELLSANGDAGWQTTRCWTRSNSYLQHGATAPFDEPSKAMTSVEVFTEEPTYGNTVFVDEIMTGRWS

[wh277]

Today, I encountered bugs when using the native AlphaFold2 to predict a complex. I suspect it might be caused by the pdb_seqres.txt file, which contains non-protein data.

[Gy-Lu]

Did the naive alphafold2 have the same question? Or, it was a different bug you met with FastFold?

[wh277]

I fixed the file, and now try FastFold again. It's my fault, the file was not downloaded by script in alphafold. sorry about it. google-deepmind/alphafold#623 (comment)

[wh277]

Now, no problem with multimer mode. There were some bad data in my old pdb_seqres.txt, after I re-downloaded the file, it works well. But I also met this: #144 (comment)