Biohub: AI for Biology, open science with Meta funding

Biohub.org - Leading the new era of AI-powered biology

Chan Zuckerberg Biohub - Wikipedia 

is a nonprofit research organization... funded by a $600 million contribution from Meta CEO and founder Mark Zuckerberg and his wife Priscilla Chan.

 “Curing All Disease by next century is too conservative" - Mark Zuckerberg - YouTube

Biohub started with an ambitious goal of curing, preventing, and managing all disease by the end of the century. A decade later, thanks to the convergence of frontier AI and biological data, that goal may have been too conservative. In this episode, Elad Gil and Sarah Guo sit down with Biohub co-founders Mark Zuckerberg and Priscilla Chan, alongside Biohub Head of Science Alex Rives. Together, they discuss Biohub’s $500 million virtual biology initiative, which integrates frontier AI with wet-lab work to build predictive world models of cells, proteins, and systems. They also talk about their newly announced open-source engine for digital protein and antibody design, ESMFold2; why Biohub is a nonprofit rather than a venture-backed startup; and how hierarchical simulations will soon allow doctors to treat patients at an individual, mechanistic level.

This episode of No Priors features Mark Zuckerberg, Priscilla Chan, and Alex Rives discussing the evolving mission of the Biohub.

Key Takeaways:

• Mission Evolution: Initially aiming to cure, prevent, and manage all disease by the end of the century, the team now views this as a conservative goal, bolstered by rapid advances in AI (1:26-2:35).
• Virtual Biology Initiative: Biohub has committed $500 million to integrate frontier AI with wet-lab biology. The goal is to move biology from a discovery-based science to an engineering-based one by building predictive, hierarchical world models of cells and proteins (4:21-5:52).
• ESMFold2 Launch: They recently open-sourced ESMFold2, a powerful protein-folding engine trained on billions of sequences. It allows researchers to design proteins and antibodies digitally, bypassing months of intensive wet-lab screening (28:04-31:29).
• The Power of Open Science: By operating as a non-profit and prioritizing open-source tools, Biohub aims to empower the entire scientific community to tackle diseases, including the long tail of rare conditions often overlooked by commercial efforts (17:24-21:10).
• Mechanistic Interpretability: A major focus is using AI to understand the 'black box' of biology. By interrogating model representations, researchers hope to uncover the mechanistic causes of diseases and predict off-target drug effects before entering clinical trials (14:22-16:58, 32:13-34:42).
• The "Virtual Cell": The next big strategic milestone is to scale these models up from individual proteins to a comprehensive digital representation of the entire cell (46:51-47:44).