A landmark study has revealed the dynamic circuits that govern how human genes work in immune cells, moving beyond DNA sequencing to show scientists exactly what happens when genes are turned on or off. Researchers from Gladstone Institutes, UC San Francisco, and Stanford University, working with Biohub, systematically tested nearly 12,800 different genes across 22 million human immune cells, creating the largest functional map of immune cell genetics ever assembled.

What Makes This Discovery Different From Previous Genomics Research?

For three decades, biology has progressed through distinct waves. The Human Genome Project gave us the blueprint of our genes. Projects like the Human Cell Atlas showed us how different cells read that blueprint. Now, scientists have entered a third wave: understanding what actually happens to cells when you make targeted genetic changes. This shift from observation to intervention represents a fundamental change in how researchers approach disease.

The team used a cutting-edge technology called Perturb-seq, which allowed them to "turn off" nearly 12,800 different genes one by one in human T cells, the critical immune cells that orchestrate how the body fights disease. Rather than relying on experimental cell lines that have long been the standard in laboratory research, the team performed massive screens on actual human immune cells from blood donors. This approach revealed how genes function in their natural state, providing a much clearer roadmap for treating autoimmune diseases and designing better cancer therapies.

"To understand the significance of this study, you have to look at the last three decades of biology. First came the Human Genome Project, which gave us the blueprint of our genes. Then, projects like the Human Cell Atlas showed us how different cells read that blueprint. Now, we're in a grand third wave: discovering what happens to cells when you make targeted changes within the genome," said Alex Marson, MD, PhD, director of the Gladstone-UCSF Institute of Genomic Immunology.

Alex Marson, MD, PhD, Director of the Gladstone-UCSF Institute of Genomic Immunology

How Does This Research Help Train AI Models for Better Predictions?

The dataset represents the largest contribution yet to the Billion Cells Project, a Biohub-led effort to generate a massive, open-source dataset of one billion single cells. This data can be used to train advanced artificial intelligence (AI) models that predict how cells behave, speed scientific discovery, and uncover new ways to treat disease. The project is part of Biohub's global Virtual Biology Initiative, which seeks to create the open-data foundation for AI-accelerated biology.

One of the key insights from the research is that AI models must be trained on diverse cell states and health scenarios to make accurate predictions. If models only learn from one context, they will struggle to predict how cells behave in the real world. The study provides proof that rich, systematic data including context-dependent responses is essential for building reliable virtual cells.

• Gene Circuit Mapping: The study reveals how genes work together to influence immune function, operating differently depending on circumstances such as whether cells are resting or fighting an infection.
• Context-Specific Data: The research demonstrates that understanding how genes function in different cellular states is essential for predicting cell behavior and designing effective treatments.
• Open-Source Availability: By making the data freely available to scientists worldwide, the research enables the development of more predictive AI models of cellular behavior and accelerates biological discovery.
• Precision Medicine Applications: The dataset provides the missing link between a person's DNA and their health, helping identify regulatory pathways through which natural genetic variants influence traits like lymphocyte counts.

"Previously, we were just cataloging individual genes. Now, we can begin to connect them into a circuit, much like the circuits that control your computer. This is fundamental to understanding how a cell actually 'thinks' and responds to its environment," explained Ronghui Zhu, PhD, a postdoctoral scholar at Gladstone and co-first author of the study.

Ronghui Zhu, PhD, Postdoctoral Scholar at Gladstone Institutes

What Are the Practical Implications for Cancer and Autoimmune Disease Treatment?

The research has immediate applications for cancer immunotherapy and autoimmune disease treatment. Scientists are already pivoting to the next phase of their work in collaboration with Weill Cancer Hub West, which will focus specifically on cancer. They plan to apply their genome-scale Perturb-seq approach to track how genetic changes alter human T cells as they infiltrate and interact with complex tumor environments.

The study also demonstrates how standardized, large-scale functional genomics datasets can serve as the foundation for the next generation of AI models in biology. By making these data open-source and freely available, researchers around the world working at the intersection of AI and biology can build more predictive models of cellular behavior and make transformative advances in precision medicine.

"We now have a fundamental rulebook of how genes control T cell responses. Our hope is that this becomes a standard lookup table for the entire field, allowing any scientist to instantly see how a specific gene affects cells of the human immune system," noted Alex Marson.

Alex Marson, MD, PhD, Director of the Gladstone-UCSF Institute of Genomic Immunology

The massive scale of the research was enabled by partnerships with industry leaders 10x Genomics and Ultima Genomics. 10x Genomics provided high-resolution single-cell analysis capabilities, while Ultima Genomics contributed high-throughput sequencing technology. Together, these platforms allowed the scientists to screen a staggering 33.4 million cells, ultimately yielding 22 million high-quality cells used for the final analysis and map.

The study, titled "Genome-Scale Perturb-seq in Primary Human CD4+ T cells Maps Context-Specific Regulators of T Cell Programs and Human Immune Traits," was published in the journal Cell on August 28, 2026. The research represents a watershed moment in immunology and genomics, moving the field from understanding what genes are to understanding what genes actually do in living cells.