Support for precomputed MSA in Colab and mmseqs in CLI #53
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Dear authors of AlphaLink2,
Thank you very much for open-sourcing your work on adding more global constraints for protein structure predictions.
There are some global constraints that I can't achieve with designing MSAs, especially for weak interactions and certain angles, where I hope AlphaLink2 will help. I hope my PI will give me good HIV next year.
I added support for precomputed/manually designed MSAs for the Colab notebook and the CLI, while working on the CLI, I also added support for the MMseqs web server, so the sequence databases are not necessary anymore. The last notable quality of live improvement is by making a pickle.gz from the csv directly in run_alphalink.sh. There behaviour does not change when running your commands.
I ran the Colab notebook and got expected results on the multimer that I tested. I added my changes to run_alphalink.sh, before letting it refactor by a LLM and then checking the code again by eye. I did not fully run the prediction, but the model started and did not complain about a missing file and the immediate files also looked as expected.
The only thing, where I am not sure is how I used homomers (e.g. here in your code and commented out here for my MMSeqs and here in my precomputed), as I always expect an MSA part, so homomers are always different MSA/entities(?).
Please let me know if there are some things that you want changed.
Happy holidays and I wish you all the best in the battle of big dinners and festivities at the end of the year!