From the perspective of performance, AlphaFold2 (and this dogma) have cracked the likely structure of various proteins. However, it is well-accepted that this dogma may not hold true for all proteins.

The low-hanging fruit was picked. Some problems remain.

Ou recent preprint: Alphafold2 and DEER biorxiv.org/content/10.110…

Happy to announce the 2024 QBioS PhD @ GT Hands on Modeling Workshop! Learn to use AI (AlphaFold2) to examine protein structure and folding! Sign up here:

sites.gatech.edu/qbios-workshop…

Due to budget cuts, this may be the last public workshop we run- sign up soon before it fill up!

There is a lot to this problem (to be covered in other 🧵's), and the problem of protein-molecule, protein-protein, and protein-drug interactions, making the usage of AlphaFold2 in real-life scenarios difficult. The functional problem extends beyond a static training database.

(🧵) New preprint: AlphaFold2-RAVE-docking for conformation selective drug discovery

AF2 fails at high quality inactive kinases,but with AF2RAVE we find classical DFG-out/A-loop folded structures for DDR1,Abl,Src where docking works like a charm arxiv.org/abs/2404.07102
Code soon

4/9 🧠vs🤖 🏆

In a realm once ruled by human genius, AI now pioneers scientific marvels.
- In '15, it decoded flatworm regeneration
- '20 saw AlphaFold2 unveil 220M+ protein structures
- '23 #GNOME was crafting the blueprint for new materials.

AlphaFold2 modeling of Brig1 showed structural homology to uracil DNA glycosylases, but with a larger predicted substrate pocket in the Brig1 model 👀 8/

A seasoned researcher in AI and biology raises important questions about the potential of large language models in addressing scientific queries.

He wonders if advancements like ChatGPT and AlphaFold2 signify a breakthrough in solving scientific mysteries.

Deep learning on computational biology and bioinformatics tutorial: from DNA to protein folding and alphafold2 | AI Summer hubs.ly/Q026lq9z0 #MachineLearning #DeepLearning #ai #ai summer #machinelearning #artificialintelligence #python

👇 from a fantastic, wise letter to Nature by Jennifer Listgarten

“DeepMind researchers performed a fantastic feat with the development of AlphaFold2, substantially moving the needle on the protein structure prediction problem. Protein structure prediction is a tremendously important…

Don't miss 'AI enthusiasm', my blog series posted on the awesome DEV community DEV Community

I talk about the impact of AI on everyday life: my last post is on AlphaFold2, by Google DeepMind, a game-changer in protein science!

dev.to/astrabert/ai-e…

#ai #blog #TECH4ALL

For instance, training AlphaFold2 relied on costly data from the Protein Data Bank.
This highlights a broader issue in scientific research, where challenges like protein structure prediction are uniquely suited for AI due to their well-defined nature and ample data availability.

🙌 Happy to share Muhammad Asif Ali and Gustavo Caetano-Anollés's latest paper:

📰 'AlphaFold2 Reveals Structural Patterns of Seasonal Haplotype Diversification in SARS-CoV-2 Spike Protein Variants'
🔦 Full text at brnw.ch/21wIT7D.

University of Illinois

The protein folding pipeline is AlphaFold2-based and consists of database search, multiple sequence alignment and AlphaFold2 evoformer and structure modules. For a set of proteins of length < 1K, Open Omics gets 17x speedup vs. prior CPU baseline on the end-to-end time. [3/n]

生成AIで遺伝子配列からタンパク質構造をかなり正確に予測できるらしい。GoogleのDeepMindが開発したAlphaFold2の性能がすごいらしい。

5/ Identification of a molecule based on density is non-trivial, we considered the density, structural predictions of all SV proteins by Alphafold2, masspec and our protein-protein interaction assay LyTHy based on BRET

The result is synaptophysin - Syp

torusengoku💙💛 3.5 Å 以上くらいのデータ・5 Å くらいまでのデータ・それ未満の低分解能データ(例えば、7 Å に AlphaFold2 のモデルを rigid body fit したようなもの)では、含まれる情報が違いすぎるし、登録方法も検証方法も異なるべきだと思うのですが、今は一括りなのがとりとしてはつらいです。

Amy Lu Gave a course at Heidelberg about structural bioinformatics last academic year including an AF2 lecture: structural-bioinformatics.netlify.app

Heavily built upon great resources already mentioned such as Nazim Bouatta's lectures. Best post for me is still this one: moalquraishi.wordpress.com/2021/07/25/the…

autism hexafluoride perry karyon nature.com/articles/s4146…

ฅNya! Here's the latest 🗞 from [ Nature Communications ]: ' Author Correction: High-throughput prediction of protein conformational distributions with subsampled AlphaFold2' 🔗: nature.com/articles/s4146…