Chan Zuckerberg Initiative

Science is at a pivotal moment, where the intersection of AI and biology has the potential to transform how biological research is done. At CZI, we saw this moment coming. We’ve been working for nearly a decade to build the tools and generate the data sets that are providing the foundation for machine learning to help power this next leap in understanding human health and disease. Now, we’ve identified four scientific grand challenges to move us closer to our goal to cure, prevent and manage all diseases in this century Here’s what we’re focused on: 1️⃣ AI-powered virtual cell models to simulate cell behavior and accelerate discovery 2️⃣ New imaging tools to visualize biology across time and scale 3️⃣ Real-time inflammation sensing to catch disease earlier 4️⃣ Immune engineering to detect and treat conditions before symptoms appear #AIinBiology #ScientificInnovation #ScienceLeadership #science #AI

Benchmarks are critical for AI to realize its full potential for accelerating biomedical discoveries. Our new tools let developers see how well their models really work, both technically and biologically, and help scientists pick the right model for their research. With seamless integration into your local environment, they will save developers time on model evaluation so they can spend more time building models that solve real biological questions. Freely available on CZI’s virtual cells platform, these benchmarking tools include vetted, high-quality tasks, metrics, and datasets from CZI, collaborators like NVIDIA Healthcare and the Allen Institute, and others in the single-cell community. CZI is leading the effort to develop and maintain these tools as a living, community-driven resource that will evolve as the field evolves, incorporating new data, refining metrics, and adapting to emerging questions in biology. By rallying the community around shared benchmarks, we’re speeding up scientific discovery and reshaping the way we study biology and understand disease.

"What you’ve come up with is a whole new world of sensors—it’s pretty extraordinary." —Eric Topol, MD, Ground Truths podcast Imagine a device that doesn’t just count steps or monitor glucose—but continuously measures inflammation in your body. Shana Kelley, President of #CZBiohubCHI, joined Eric to discuss how this technology—and other Biohub efforts—could open the door to a new era of personalized medicine. Listen to the episode (and hear Eric volunteer to be the first tester): https://lnkd.in/gzpSQM9E

Congratulations to David Fajgenbaum, MD, MBA, MSc on being named to the 2025 TIME100 Next list! His leadership in harnessing AI to uncover new uses for existing medicines is transforming what’s possible for patients and families facing rare disease 🙌

👋🔬 Meet Sjoukje van der Stegen of the New York Biohub! Van der Stegen imagines a world where immunotherapy is as accessible as any prescription drug. As lead of the Immune Cell Engineering and Development Platform at the New York Biohub, she is working to generate engineered immune cells from induced pluripotent stem cells, or iPSCs, for a variety of uses, including immunotherapies. Read more about van der Stegen’s immune system research ⤵️ https://lnkd.in/gcFFQR6B #science #cancer #immunotherapy #iPSC #ImmuneSystem #CZBiohubNY

🚀 New CZI/Biohub Funding Opportunity: Live Tissue Spatiotemporal Omics Current spatial omics technologies have opened windows into the proteome and metabolome, but there’s a critical limitation: they typically rely on frozen or fixed tissue sections. This makes it nearly impossible to capture dynamic physiological processes as they unfold. Without the ability to monitor molecular changes in real time, we miss key opportunities to understand evolving biology and investigate disease mechanisms. Our new RFA will address this unmet need. We’re calling for innovative, biocompatible, minimally invasive sampling and analytical technologies that push beyond today’s barriers. The goal: enable real-time, high-resolution interrogation of proteomic and metabolomic landscapes in live tissues. 🔬 With a focus on immunology and autoimmune disease, this program seeks scalable, engineering-driven solutions that can transform both ex vivo and in vivo research. We are looking for proposals that align closely with the mission of the Chicago Biohub and our work on inflammmation. If you’re working at the cutting edge of bioengineering, omic-level molecular analysis, or immunology, this is your chance to help catalyze the next wave of breakthroughs in spatial omics. 👉 Learn more and apply: https://lnkd.in/ehxbK5qC

Sepsis, autoimmune flares, cardiac events, they are all driven by inflammation that often goes unseen until it’s too late. Real-time protein monitoring gives us a chance to catch it early and act fast, before the damage is done. Learn how implantable sensors for continuous protein monitoring could transform medicine, from #CZBiohubCHI President Shana Kelley: http://czi.co/4mtxvbb.

“We’re committed to building open, public AI infrastructure that reflects how students learn and makes the best of learning science accessible for students and educators everywhere.” —Sandra Liu Huang, Head of Education at CZI For more than a decade, we’ve partnered with educators, researchers, and developers to expand what’s possible for every learner. 🎉 Today, we’re sharing an early public release of Knowledge Graph and Evaluators, as part of our efforts to scale proven teaching and learning practices to benefit every learner. As part of this launch, we are providing early access to: 🔹 Knowledge Graph—connected high-quality education datasets, including academic state standards, math learning components, and math learning progressions for the development of rigorous education tools. This is also available as a Connector in Anthropic’s Claude, so educators can work directly with these education datasets to help align content across standards and develop more precise lesson plans and problem sets to support their teaching practices. 🔹 Evaluators—open-source models that help assess outputs to ensure AI-generated content is accurate, rigorous, and trustworthy. As part of this early release of our tools, we are now calling our work in education Learning Commons—a name that reflects our sharpened focus and role within the larger education ecosystem. More from Sandra:  http://czi.co/3IxNQxy #LearningCommons #EdTech #AI #Education #EducationTechnology

How do you accurately track cells in 3D bioimaging data? It’s a question that has challenged researchers for years. Cells move, divide, and overlap in complex ways that make accurate tracking one of the toughest problems in bioimaging. #CZBiohubSF researchers built Ultrack, an AI tool that can follow every cell in 3D over time — from a few cells in a dish to whole embryos. By reducing errors and speeding up analysis, Ultrack is already helping scientists study development, immunity and disease more efficiently. This work points to new possibilities for large-scale imaging and cell tracking — giving researchers a clearer view of how cells behave in health and disease, and helping lay the groundwork for new ways to prevent and treat illness. Read more: http://czi.co/47kmk0I #OpenScience #AI #Microscopy #Science

“I am so excited about CZI’s Imaging the Future Conference! Patient-centered research is all about using the best tools at our disposal to improve lives. When patients guide researchers to a meaningful field of view, advanced imaging can accelerate our understanding of rare diseases and our search for treatments.” We think that our partner Dylan Verden, PhD from KIF1A.ORG really captured the vibe at our #ImagingtheFuture Conference. A huge thank you to everyone for attending and helping create the connections that are pushing science forward to improve human health. #AI #imaging #OpenScience #science