Can AlphaFold v2 predict the structure of a tetramer?

Question

[Apprentice]

I would like to predict a tetramer structure based on 4 fasta sequences. Is AlphaFold v2.1 capable of predicting the tetramer structure?

Answer 1

[Mensur Dlakic]

Yes. You can find more details by Googling, or from the links below.

• https://colab.research.google.com/github/deepmind/alphafold/blob/main/notebooks/AlphaFold.ipynb
• https://deepmind.com/research/publications/2021/protein-complex-prediction-with-alphafold-multimer
• https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1
• https://github.com/deepmind/alphafold

[Dunois]

Also worth taking a look at ColabFold's table here: https://github.com/sokrypton/ColabFold#making-protein-folding-accessible-to-all-via-google-colab.

[Apprentice]

Thank you for your thoughtful response. I have confirmed that it is possible to estimate the structure of multimer as you suggested.

Is it correct that this multimer structure estimation function is possible not only for dimers but also for tetramers and more?

[Dunois]

I think they tested it with dimers and trimers, but nothing more complex than that. You're just going to have to try it out for yourself. I suspect the model accuracy will be pretty low for anything involving more than three subunits. (Someone can correct me if I'm mistaken.)