AlphaFold Unofficial

@alphafold.bsky.social

Unofficial account exploring the intersection of biology, molecules, science, AI and protein folding with AlphaFold.

🧬 Figure Friday: Celebrating discoveries from the Protein Science community Can AI predict how proteins assemble? Ben Corry and colleagues show that AlphaFold can accurately predict the oligomeric states of many proteins, while also highlighting cases where predictions remain challenging.

Bild

John Jumper, the Nobel laureate behind AlphaFold, is leaving Google DeepMind after ~9 years for Anthropic. AlphaFold predicted 200M+ protein structures. Lands amid the talent war, days after Shazeer left Google for OpenAI. (Disclosure: we run on Anthropic's Claude.)

John Jumper, the Nobel laureate behind AlphaFold, is leaving Google DeepMind for Anthropic

John Jumper, who shared the 2024 Nobel Prize in Chemistry for AlphaFold, is leaving Google DeepMind after nearly nine years to join Anthropic. AlphaFold predicted 200M+ protein structures. The move lands amid the AI talent war, days after Shazeer left Google for OpenAI.

zubnet.ai

Will AlphaFold et al. figure out how the genome works? Doubtful, says @philipcball.bsky.social. I think he's absolutely right. The genome is a complex adaptive system, w/ many layers of regulation and buffering—a "sensitive organ of the cell," said Barbara McClintock—in 1983!

Why the Human Genome’s Tangled Physicality May Confound AI | Quanta Magazine

Our genetic heritage is not a blueprint or an algorithm, as many biologists have imagined, but something else entirely.

quantamagazine.org

A tough week for Google’s AI efforts. Along with losing the co-lead of Gemini to OpenAI, they are also losing the Nobel Prize winning lead of the AlphaFold project at DeepMind to Anthropic. He won a Nobel Prize in Chemistry in 2024 for using AI to predict 3D protein structures.

Bild

I’m not anti-tech or investment in tech. I’m excited about solar and AlphaFold and ocean monitoring. I love science; and I know the less flashy sides of science are suffering too. I just don’t want to see the humanities, our collective history, gutted to the point where it’s too late.

🔬🚀 It's out! Run 100s of imaging experiments before touching a microscope. #VLab4Mic reproduces them in silico from PDB/AlphaFold to help you image right first time. From rockstars Damián Martínez & Bruno Saraiva, inaugural preprint of @mariodelr.bsky.social new lab www.biorxiv.org/content/10.6...

Mario Del Rosario@mariodelr.bsky.social · 2mo ago

Congratulations to Damián Martinez, Bruno Saraiva, @taylashakespeare.bsky.social, @wmarkbates.bsky.social, Dylan Owen, @christlet.bsky.social and @henriqueslab.bsky.social We hope VLab4Mic helps researchers design better experiments and ask better biological questions 💻 vlab4mic.henriqueslab.org

It always gets me the lengths people go to try to argue that LLMs and image generation models "aren't really AI", and then they point to "AI" things, and they're based on the same underlying tech. AlphaFold is built on latents, diffusion, *and* Transformers.

Before AlphaFold: Biology was like trying to understand machines without seeing their shapes. After AlphaFold: Scientists suddenly received a massive 3D blueprint library for life’s molecular machinery. This is how we can use AI for good. But w/o regulation, we risk harm to society.

AlphaFold: Five Years of Impact

Explore five years of AlphaFold’s impact on biology. Learn how this Nobel Prize-winning AI is accelerating scientific discovery globally

deepmind.google