The sequence in design.fasta is different from the sequence in best.pdb

sokrypton / ColabDesign

Making Protein Design accessible to all via Google Colab!

549 stars 127 forks source link

The sequence in design.fasta is different from the sequence in best.pdb #124

Closed Lyang556 closed 1 year ago

Lyang556 commented 1 year ago

I am trying to use RFdiffusion in ColabDesign for protein design, and I found that the sequence in design.fasta is different from the sequence in best.pdb, is this normal? Below is one of my results

design.fasta: TYKVVALFTG-----/---- best.pdb: TYKVVALFTGC-----/----

represents other amino acids You can see that the sequence of best.pdb has one more C than the sequence of design.fasta

Lyang556 commented 1 year ago

The --- represents other amino acids and the / represents the chain break

sokrypton commented 1 year ago

That shouldn't be the case. Can you share an example input you used?

Lyang556 commented 1 year ago

@sokrypton Here is an example input. contigs:10-17/A1008-1031/30-38/0 C22-83 pdb:5zng copies:1 initial_guess:True

Lyang556 commented 1 year ago

The other parameters in the Colab RFdiffusion are default.

sokrypton commented 1 year ago

Thanks for reporting the bug! I've fixed this in the updated version. Tell me if you still see any issues