Sergey O
Sergey O
Please try again! The error should be fixed now.
Thanks for the reports... Still trying to debug why this is happening... One quick solution is to run with CPU only. For some reason, GPU version is having some issues...
you can use our advanced notebook for this! https://colab.research.google.com/github/sokrypton/ColabDesign/blob/gamma/af/examples/predict.ipynb msa_method=single_sequence template_mode=custom If the sequence of the template matches the query, and no msa is used, then the output should copy...
In your case, just set copies=4, propagate_to_copies=False and list all chains from the pdb out. See end of notebook for instructions.
Can you share a screenshot of the template features (should appear after prep_inputs cells)? 1) if you say "tetramer" it will try download a protein by name of "tetramer" from...
Looks like only features for chain A were loaded. Did you set: chain="A,B"?
Thanks for tagging me! I'll take a look. For context: I've implemented a version that works with keras and model.fit() here: https://github.com/sokrypton/AccAdam_TF2
Remove .git Instead of `[email protected]` try `[email protected]`
The easiest way would be to modify the input bias. af_model._inputs["bias"] is a (length,20) matrix. By default it is all zeros. If you set some positions and amino acids to...
Thanks for the report, the issue has been fixed in v1.1.1. But if you want to update your copy see: https://github.com/sokrypton/ColabDesign/commit/aa3ced63542d61b1b5ef2b58371b90e22f11cdde