The generated pdb file

[Layne-Huang]

Hi,

Thank you so much for releasing your wonderful work. I have some questions about the generated files.
For the codesign.fa, why does the generated seq consist of only Alanine?

Besides, what is the meaning of seqs/sample.fa? Are these sequences generated by multiflow?

The sample.pdb also only consists of Alannie too.

[jasonkyuyim]

What command are you running? python -W ignore multiflow/experiments/inference_se3_flows.py -cn inference_unconditional is giving me valid sequences.

>codesign_seq_1
AEKLKEIVKKIKEIAKELDEEEKEIAEKLAKVLEKLAENEEAVKELEKFAEEVEKKLEEILKKIGCKRTLDENGKLVVES
KPASPEQLKEFKEELKKLREEGKKKLEEVFEKANLSEEEKEFAKKAGAGALLYYLSKIEGMEEVLKEFEEELKKKYSPEE
LEEIKKAGKVLAKILEKLNKEDPELALEFEKIVLALGLGA

Yes codesign_seq_1 is supposed to be the codesigned sequence.

[Layne-Huang]

Hi,

I ran the same command as you used, including the weights and the configuration file that you provided. The only argument that I change is that I set samples_per_length as 1. Could you please help me check it again?

[Layne-Huang]

I have just run the command again. The model also only generates the all-A sequences.