NECSI Resources

 International Conference on Complex Systems (ICCS2007)

A Modular Gene Regulatory Network Model of Artificial Ontogenesis

Amer Ghanem
Department of Computer Science, University of Cincinnati

Ali Minai
Department of Electrical and Computer Engineering, University of Cincinnati

     Full text: Not available
     Last modified: October 16, 2007

Abstract
All but the simplest multicellular organisms are highly
heterogeneous, with specific arrangements of organs and structures
central to their functionality. At the level of individual
organisms, this organization must be extremely robust against
variations in the developmental process and in the environment. At
the same time, evolution requires sufficient variation {\it across}
organisms in order for selective pressures to work. Thus, evolving
systems face a critical tension: How to maintain a stable
heterogeneous organization while allowing sufficient phenotypic
variation.

Recent work in evolutionary developmental biology has demonstrated
that modularity is the key enabler for balancing variation and
stability in animals. The deployment of the same developmental
modules in different spatial and temporal combinations (called
heterotopy and heterochrony, respectively) can generate a wide range
of forms without the need for much change within the modules. Thus,
useful -- and sometimes critical -- features evolved with great
difficulty can be conserved as modules, while large changes in the
regulatory mechanisms underlying their use can provide a large part
of the variation needed for evolution.

In this paper, we combine the idea of regulatory variation in a
modular developmental system with Kauffman's hypothesis of cell
types as attractors. We present a simple model for the ontogenesis
of two-dimensional shapes where an attractor network of regulatory
genes robustly controls the deployment of genetic network modules
that determine phenotypic features. We use this model to show that a
wide variety of forms can be generated purely through changes in the
connectivity and weights of the regulatory network while conserving
the rest of the system. We also investigate the effects of varying
switching conditions and initial asymmetries in maternal proteins.







Maintained by NECSI Webmaster Copyright © 2000-2007 New England Complex Systems Institute. All rights reserved.