According to the latest news on the website of the American Association for the Advancement of Science, American scientists have utilized the powerful computing power of the world's top supercomputers to build the largest and most detailed animal brain simulation system to date. This virtual model fully reproduces the structure and function of the mouse cerebral cortex, including nearly 10 million neurons, 26 billion synapses, and 86 interconnected brain regions, becoming a new platform for studying brain operation mechanisms. This also means that scientists are standing at a brand new starting point - gradually evolving from understanding the brain to building it. This breakthrough achievement was achieved based on the Japanese supercomputer "Fuyue". Fuyue can perform trillions of operations per second and has the ability to handle massive amounts of data and complex simulation tasks. This project is led by the Allen Institute for Brain Science in the United States and the University of Electric Communications in Japan, and is jointly completed by three Japanese institutions. The research team utilized real neurobiological data from the "Allen Cell Type Database" and "Allen Connection Graph" provided by the Allen Institute for Neuroscience to provide precise biophysical foundations and structural blueprints for the virtual brain. Then, using the brain modeling toolkit independently developed by the Allen Institute, these data are transformed into a dynamically running digital cortex model. During the simulation process, the specialized neuron simulator "Neulite" converts mathematical equations into neurons with real biological behavior. These virtual neurons can generate electrical pulses, transmit signals, and form dynamic networks like living cells. The entire simulation process is highly realistic, not only reproducing the dendritic branch structure of neurons, but also fully presenting the signal transmission process between synapses and the fluctuation of cell membrane potential, as if observing real brain tissue activity in real time. Scientists can now use this model to explore brain mechanisms in an unprecedented way. They can simulate neurological diseases such as Alzheimer's disease and epilepsy in a virtual environment, track how lesions spread in neural networks, and study the formation mechanism of brain waves, the neural basis of attention, or the propagation path of epileptic seizures. In the past, these issues could only be validated through one animal experiment at a time, which was time-consuming and difficult to replicate. Nowadays, scientists can quickly propose hypotheses and repeatedly test them in the digital brain, greatly improving research efficiency. This achievement provides a new tool for understanding the neural basis of cognition and consciousness, and is expected to reveal early changes in brain diseases before symptoms appear, evaluate potential therapies, and accelerate the development of new drugs. The team stated that although this is a significant progress, it is still only the first step towards whole brain simulation. The real challenge lies in the details, only by fully reproducing the complexity at the biophysical level can the model have more scientific value. The long-term goal of the team is to move towards the digital reconstruction of the human brain. (New Society)