AlphaFold2 predictions are less efficient with the latest version

[cdelam]

Expected Behavior

The predictions are worse with the latest version of sokrypton/ColabFold In the past I used the following link, but it is getting an error now: https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/beta/AlphaFold2_advanced.ipynb Currently I'm using the link (merge beta into main): https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb

Current Behavior

More or less the same prediction between the two sites. In fact, the result with the beta site corresponds to a pdb 6sdz file from Proline 11 to Lys35, which is proof of the effectiveness of the beta version.

Steps to Reproduce (for bugs)

ColabFold Output (for bugs)

File enclosed containing the same prediction on the two sites

Context

Providing context helps us come up with a solution and improve our documentation for the future.

Your Environment

Results.pdf