Theoretical Biology Laboratory

Outline

We study biological phenomena using theoretical methods. The progress of modern biology is causing a rapid increase in information about molecular level phenomena. We now understand that many biological phenomena are governed by complex regulatory systems including molecules, cells and organs, and that biological functions emerge from the whole of the systems. We adopt mathematical or computational methods to decipher huge amounts of experimental information, and to give integrative understanding of complex biological systems. For example, we recently developed a theory to understand the dynamics of activities of bio-molecules from the structure of regulatory networks. Using this theory, we can reduce complex networks including more than 100 molecular species to smaller networks with a few molecules without any loss of the dynamic potential to produce steady states. In addition to the theory of networks, we study thermotaxis in C. elegans, pattern formation of neuronal dendrites and other biological phenomena using mathematical modeling. We also have multiple projects of collaboration with experimental biologists. Our final goal is to open a new bioscience that will progress by repeating theoretical predictions and experimental verification.

Recent Research Topic

Structure of regulatory networks and dynamics of bio-molecules

Fig. 1

The structure of a network specifies the incompatible region. The incompatible region determines the maximum number of possible steady states.

Fig. 2 Analysis of the signal transduction pathway of an EGF receptor

There are only five informative nodes, which act as memory circuits.

Regulation of biological molecules constitutes complex network systems, which generate various developmental and physiological functions via dynamics of molecular activities. However, we do not understand the significance of the complex structure of regulatory networks in these dynamics. In other words, very little is known about the relationship between the structure of a network and its dynamic nature. Recently we have developed theories to understand dynamic properties only from the structure of regulatory networks. The basic premise is functionality: the dynamics of an activity of a molecule should be uniquely determined by the activities of controlling entities in the network. We formalize two aspects of functionality, incompatibility and independency. The incompatibility determines the upper limit of the number of steady states of molecular activities realized by a given network. The independency determines the possible combinations of states of the system. These constraints (upper limit and possible combinations) on the molecular activities are determined only from the topological structure of a network without any assumptions of dynamics. Thus, inconsistency with the experimental data leads to a prediction of unknown states or unknown regulations. This method was applied to some experimentally determined regulatory networks, including the gene network for early development of ascidians, and the network of signal transduction pathways. Some of these networks include more than 100 molecules. We found a small number of molecules whose activity is responsible for the diverse steady states. We are collaborating with groups of experimental biologists to measure the diversity of activities of these informative molecules. The inconsistency with the observed expression pattern indicated the presence of unknown regulation.

Mathematical modeling of gene expression of vertebrate segmentation

Segmentation in the vertebrate PSM (presomitic mesoderm) is established by a series of pattern formations through the dynamics of gene expression at different levels. Some downstream genes suppress the activity of upstream genes, the negative feedback of which seems to bring about transient dynamics in patterning. We have developed two mathematical models in which the negative-regulator genes are different. We found that the previously accepted regulation model cannot explain the mutant expression patterns whereas the newly proposed model can. The mathematical models provide an integrative understanding and a working hypothesis for a regulatory network system including many genes. This study was done in collaboration with Drs. Takada and Takahashi of the National Institute for Basic Biology.

Fig. 3 Two models for the regulations of segmentation genes

Left: Model A (Previously believed), Right: Model B (Our hypothesis).

Fig. 4

Dynamics of gene expressions for Ripply null-mutant are completely different between the two models. Left: Model A (Previously believed), Right: Model B (Our hypothesis). Only model B can explain the experimental results.

Fig. 1

Reproduced, with permission, from A. Mochizuki, Structure of regulatory networks and diversity of gene expression patterns, J. Theor. Biol. 2008, 250, 307. © (2012) Elsevier

Fig. 3

Reproduced, with permission, from J. Takahashi, et al. Analysis of Ripply1/2-deficient mouse embryos reveals a mechanism underlying the rostrocaudal patterning within a somite, Dev. Biol. 2010, 342, 134. © (2012) Elsevier