Sergey O

This issue should be resolved now. It seems conda installation break google-colab %shell magic functions. Replacing this function with just os.system() seems to do the trick.

The best result is saved during the 3rd stage (when we switch to one_hot). For some very difficult targets, there might not exist a one_hot solution with < 0.5 rmsd......

Wow... This does look like a difficult target! You need 6 disulfides to hold it together: Unconstrained design looks like this (maybe if we increase number of iterations at stage...

I think it's a great idea... but i'm a bit concerned about the terminology used here. Calling this "inverse folding" is misleading. Inverse folding has a very specific meaning (you...

I would suggest renaming this document to "protein design" and grouping the design methods based on the probabilities they are approximating. P(sequence) = various language models etc. P(sequence | structure)...

Mind if I accept and then rename to `proteindesign.md` ? 😄

Hi guolinke, will it run in uni-fold-jax?

Thanks for offer to help running Uni-Fold on Colab! My first attempt at installing Uni-Core failed :( https://colab.research.google.com/drive/171kK3kCsAyPf3vXi4PcVrN91Af_tQjfk if I comment out this line. the installation continues: `check_cuda_torch_binary_vs_bare_metal(torch.utils.cpp_extension.CUDA_HOME)` But it...

@ZiyaoLi does that mean `convert_unifold_to_alphafold.py` was created somewhere? :D

I believe, if you wrap the K.in_train_phase() function with a keras.layers.Lambda() function, that should fix the bug. I noticed in new tensorflow version, when in eager mode, the train mode...