arXiv subjects

Shuji Ishihara

Publications and source records attributed to Shuji Ishihara.

24 records · Page 2Linked to original sources

arXiv2008

Network Evolution of Body Plans

Segmentation in arthropod embryogenesis represents a well-known example of body plan diversity. Striped patterns of gene expression that lead to the future body segments appear simultaneously or sequentially in long and short germ-band development, respectively. Regulatory genes relevant for stripe formation are evolutionarily conserved among arthropods, therefore the differences in the observed traits are thought to have originated from how the genes are wired. To reveal the basic differences in the network structure, we have numerically evolved hundreds of gene regulatory networks that produce striped patterns of gene expression. By analyzing the topologies of the generated networks, we show that the characteristics of stripe formation in long and short germ-band development are determined by Feed-Forward Loops (FFLs) and negative Feed-Back Loops (FBLs) respectively. Network architectures, gene expression patterns and knockout responses exhibited by the artificially evolved networks agree with those reported in the fly Drosophila melanogaster and the beetle Tribolium castaneum. For other arthropod species, principal network architectures that remain largely unknown are predicted.

q-bio.MN↗

arXiv2006

On the transient and steady state of mass-conserved reaction diffusion systems

Reaction diffusion systems with Turing instability and mass conservation are studied. In such systems, abrupt decays of stripes follow quasi-stationary states in sequence. At steady state, the distance between stripes is much longer than that estimated by linear stability analysis at a homogeneous state given by alternative stability conditions. We show that there exist systems in which a one-stripe pattern is solely steady state for an arbitrary size of the systems. The applicability to cell biology is discussed.

nlin.PS↗

arXiv2006

A mass conserved reaction-diffusion system captures properties of cell polarity

Various molecules exclusively accumulate at the front or back of migrating eukaryotic cells in response to a shallow gradient of extracellular signals. Directional sensing and signal amplification highlight the essential properties in the migrating cells, known as cell polarity. In addition to these, such properties of cell polarity involve unique determination of migrating direction (uniqueness of axis) and localized gradient sensing at the front edge (localization of sensitivity), both of which may be required for smooth migration. Here we provide the mass conservation system based on the reaction-diffusion system with two components, where the mass of the two components is always conserved. Using two models belonging to this mass conservation system, we demonstrate through both numerical simulation and analytical approximations that the spatial pattern with a single peak (uniqueness of axis) can be generally observed and that the existent peak senses a gradient of parameters at the peak position, which guides the movement of the peak. We extended this system with multiple components, and we developed a multiple-component model in which cross-talk between members of the Rho family of small GTPases is involved. This model also exhibits the essential properties of the two models with two components. Thus, the mass conservation system shows properties similar to those of cell polarity, such as uniqueness of axis and localization of sensitivity, in addition to directional sensing and signal amplification.

q-bio.CB↗

arXiv2005

Turing Pattern with Proportion Preservation

Although Turing pattern is one of the most universal mechanisms for pattern formation, in its standard model the number of stripes changes with the system size, since the wavelength of the pattern is invariant: It fails to preserve the proportionality of the pattern, i.e., the ratio of the wavelength to the size, that is often required in biological morphogeneis. To get over this problem, we show that the Turing pattern can preserve proportionality by introducing a catalytic chemical whose concentration depends on the system size. Several plausible mechanisms for such size dependence of the concentration are discussed. Following this general discussion, two models are studied in which arising Turing patterns indeed preserve the proportionality. Relevance of the present mechanism to biological morphogenesis is discussed from the viewpoint of its generality, robustness, and evolutionary accessibility.

q-bio.CB↗

arXiv2004

Magic number 7 $\pm$ 2 in networks of threshold dynamics

Information processing by random feed-forward networks consisting of units with sigmoidal input-output response is studied by focusing on the dependence of its outputs on the number of parallel paths M. It is found that the system leads to a combination of on/off outputs when $M \lesssim 7$, while for $M \gtrsim 7$, chaotic dynamics arises, resulting in a continuous distribution of outputs. This universality of the critical number $M \sim 7$ is explained by combinatorial explosion, i.e., dominance of factorial over exponential increase. Relevance of the result to the psychological magic number $7 \pm 2$ is briefly discussed.

nlin.CD↗

arXiv2001

Simple dynamical model with history dependence for a sandpile experiment

A lattice dynamics model is proposed for the history dependence observed in sandpile experiments. The dependence of the stress distribution on the preparation of the sandpile is explained as a dependence of certain attractors on the preparation of the system. The model has three phases, but the history dependence is shown to exist only in the phase where a perturbation is amplified selectively rather than globally when propagating in the downflow direction. The condition for this history dependence is given in terms of the spatial Lyapunov exponent.

cond-mat.soft↗