Tareq Hameduh

AlphaFold’s success builds on decades of tech advances, massive efforts to collect, curate, and share structural data, and rigorous benchmarking. What will it take to see a similar leap in structural ensemble prediction? @stephanieaw.bsky.social and I share some thoughts here 👇 Feedback is welcome!

Anyhow, thanks for your efforts.

That would be awesome.

However, do you think it would be better to do 5000 sample compared to 1000 sample?

And what do you think would be faster to run it on a local machine with RTX 4090 or keep on the colab with the A100?

Fourth, facilitating the exploration of potential conformations before deciding which to simulate further so instead of simulations that takes weeks you will do the modelling of Bioemu than decide what to do next,

Fifth, it can compare the wild versus mutant sequence configurations

Second it can help you to scan the preferred conformation spaces of protein sequence to show you if there is a preferred configuration or multiple configurations,

Third it can shows the most unstable part of protein so you can investigate there if needed,

Hello Pablo,

First I need to thank and your team for this cool technology

And I like to summarise the uses I can see

it’s a great tool to check the stability of your structure (protein or peptide) as a first point,

where we prepose a "proteomic code" derived from DNA's genetic code, suggesting that specific amino acid pairings determine protein folding (intra-connections) and protein-protein interactions (inter-connections) based on DNA and biophysics.

The proteomic code: Novel amino acid residue pairing models “encode” protein folding and protein-protein interactions Recent advances in protein 3D structure prediction using deep learning have focused on the importance of amino acid residue-residue connections (i.e.,…

Our new published paper,

The proteomic code: Novel amino acid residue pairing models “encode” protein folding and protein-protein interactions:

www.sciencedirect.com/science/arti...