Cells are the building blocks of all multicellular organisms, and all cells and their functions are realized through chemical reactions. A group of diverse molecules confined in the microenvironment of a cell exhibits highly characteristic behaviors. In this theme, we will develop a mathematical theory of cellular chemical reaction systems based on probability theory, and test the theory using quantitative data to answer the questions: How can we describe phenomena at the cellular level? How do the diverse but small numbers of molecules affect cellular functions? What are the physical laws that constrain equilibrium and nonequilibrium systems consisting of a small number of molecules?

Biological systems, from organisms to cells, actively acquire and process information from the environment to determine responses such as locomotion and state switching. However, the chemical reactions that make up microscopic cells are extremely stochastic and noisy. How do cells process information using noisy chemical reactions, and how do they utilize this information? The principle is still unclear. Based on information theory and information thermodynamics, we are developing a mathematical theory for the recognition and exploration of dynamically changing environments. By combining these theories with quantitative measurements, we are trying to understand biological information processing from the viewpoint of information.

Living systems have the ability to flexibly adapt to a stochastically fluctuating environment. Darwinian evolution based on natural selection is one of the basic mechanisms of environmental adaptation. By generating genotypic and phenotypic diversity within a population, living organisms can spread the risk of unknown environmental changes, thereby increasing the probability of survival and the degree of adaptation. On the other hand, biological systems have developed brain-like organs that can proactively sense and predict the environment and select adaptive states in advance. How do these two adaptive mechanisms relate to each other? We are working on theoretical integration of these two adaptive mechanisms using the information-theoretic variational structure common to Darwinian natural selection and predictive information processing, and on the construction and application of a unified theory of adaptation in living organisms.

Single-celled organisms such as E. coli, yeast, slime mold, and cultured cells are good model systems for discovering quantitative laws in living systems. By collaborating with various experimental researchers, we combine a variety of quantitative data with various mathematical and data analysis methods to discover new laws. In particular, we are focusing on how the behavior and functions of cell populations are realized as a result of the behavior at the single-cell level and the stochasticity and diversity of each cell.

From cells to organisms, living things can recognize obtain various external and internal information via complex chemical mixtures and their dynamics. A wide variety of information is encoded by the combination of chemical substances, and the mechanism by which this information is decoded is still elusive. We work on this problem by focusing on the immune system and olfactory system, which can handle an extremely wide variety of chemical information in the body. As a target of informatics, chemical information processing is the third but unexplored subject, following visual information and auditory information. We aim to solve this problem by developing new mathematical theories and information technologies.