R-CCS Cafe 特別版 Talk1

Our daily activities from life maintenance to higher cognitive functions are all mediated by the brain. The human brain is a huge and complex network composed of about 86 billion neurons connected from each other via junctions called synapses. Although how the brain works has not been fully understood, it is possible to build a computational model of the human brain through mathematical modeling of neurons and synapses, and large-scale simulation of the entire network dynamics. Such brain simulation could, in principle, replicate various functions that emerge from the human brain. In Program for Promoting Research on the Supercomputer Fugaku, we have been working on building a human-scale brain model and a mouse-scale brain-body model. These models are developed based on known anatomical and physiological data, which were obtained from a number of different animals / subjects. Thus, these models are regarded "averaged" models. On the other hand, recent advances of experimental methods allow us to obtain massive amount of data from one single animal / subject, which will make building a digital copy of the brain for an individual subject possible. In this seminar, I will review the experimental methods that enables to build a brain digital twin, and discuss the potential impacts on basic neuroscience, precision medicine, and even society.