Using Japan’s premier supercomputer, Fugaku, scientists have created one of the most detailed and biologically realistic simulations of a brain ever built—the entire mouse cortex. This digital model includes nearly ten million neurons, 26 billion synapses, and 86 interconnected brain regions, replicating both the structure and dynamic activity of a living brain at an unprecedented level of detail. The simulation allows researchers to study brain function, test theories about cognition and consciousness, and model neurological disorders such as Alzheimer’s and epilepsy in a virtual environment. The project was led by a collaboration between the Allen Institute for Brain Science and Tadashi Yamazaki, Ph.D., from the University of Electro-Communications in Japan, along with three other Japanese research organizations. The full results will be presented at SC25, the world’s leading supercomputing conference, in mid-November. Fugaku, developed by RIKEN and Fujitsu, is one of the fastest supercomputers in the world, capable of over 400 quadrillion calculations per second. Its immense power was essential for handling the complexity of simulating a full cortical network. The system consists of 158,976 individual processing nodes, organized into racks and shelves, enabling it to process vast amounts of data and run highly detailed models. The team used data from the Allen Institute’s Brain Modeling ToolKit, including information from the Allen Cell Types Database and the Allen Connectivity Atlas, to build a biologically accurate foundation. A specialized tool called Neulite then transformed mathematical models of neurons into virtual cells that can spike, communicate, and respond to stimuli just like real neurons. The result is a dynamic, living digital cortex that mimics the electrical and chemical activity of a real brain. This simulation opens new possibilities for neuroscience. Researchers can now test how brain injuries spread, how seizures propagate through neural circuits, and how brain waves influence attention—all in a controlled, digital space. Unlike traditional methods that rely on animal studies or post-mortem tissue, this model allows for rapid, repeatable experiments and the exploration of conditions before symptoms appear. Anton Arkhipov, Ph.D., an investigator at the Allen Institute, called the achievement a major technical milestone. “This shows the door is open. We can run these kinds of brain simulations effectively with enough computing power,” he said. “It’s a technical feat, but it’s only the first step.” Yamazaki emphasized the importance of detail. “God is in the details, so in the biophysically detailed models, I believe,” he said. The team’s long-term goal is to scale up to whole-brain models, eventually including human brains, using the full range of biological data being uncovered. The project was made possible by a global team including Laura Green, Ph.D., Beatriz Herrera, Ph.D., Kael Dai, B.Sc., Rin Kuriyama, M.Sc., and Kaaya Akira, Ph.D. As supercomputing power continues to grow, the dream of building a complete, accurate digital brain is no longer science fiction—it is becoming a scientific reality.