Morphogenesis is assumed to be a dynamical system or process that has emerged from the complex interactions of many genes and structures. These systems of evolutionary transformation of morphology may either emerge spontaneously by self-organisation, or may be the product of natural selection. Nowadays authors have begun to use hybrid systems which model both self-organising and selective mechanisms of biological systems. Following Dellaert and Beer (1994a,b), Kitano (1994), Wagner and Altenberg (1995), we will study, from the point of view of self-organisation, the evolution of the genotype to phenotype map via the creation of new genes.
There are already of the new developments in GAs (genetic algorithms) that have to do with the inclusion of developmental stages between the genotype and phenotype, in other words, the creation or simulation of some artificial morphogenesis (Rocha, 1995). The authors both understand and appreciate the importance of incorporating a developmental process into schemes for the genotype to phenotype transformation. The idea has been to encode rules that will themselves self-organize to produce a phenotype, rather than the direct encoding of the phenotype itself. Such schemes bring together the two most important aspects of evolutionary systems: self-organisation and evolution (Rocha, 1995). Really, there is no direct mapping between genotype and phenotype in biological morphogenesis. Rather, biological morphogenesis is the result of a co-operative self-regulating process that is controlled by the genome. Intrinsic properties of these developmental processes, when used in conjunction with GAs, is believed to lead to much more complex morphologies than those achievable with direct mapping.
On the other hand, many biologically realistic models of different developmental processes are found in the theoretical biology literature. However, none of these more complex models have yet been used in conjunction with genetic algorithms (Dellaert & Beer, 1994a).
This paper describes our first steps toward a model of evolutionary growth of the Drosophila segmentation gene net. The identification of controller genes has been a significant recent finding in developmental biology. Networks and cascades of controller genes serve to orchestrate expression of the genome during embryo development. Now we have a lot of knowledge about mechanisms of appearance and maintenance of patterns of the controller genes expression. The network's activity as kind of self-organising mechanism of morphogenesis allows the conditions of the effectiveness of natural selection to be better understood. Stabilising and driving selection applied to these self-organising mechanisms can dramatically accelerate evolutionary complexification of developmental processes. According to this approach, self-organisation becomes apparent in evolution as self-assemblage of the gene nets and cascades.
I use up to date knowledge about structure, function and evolution of real gene networks for the purposes of computer simulation of the self-organisation of networks during evolution. Perhaps the most obvious opportunity for extending genetic algorithms is to the study of evolution itself (Mitchell & Forrest, 1995). It is well known that ideas from evolution have provided inspiration for developing these interesting computational techniques. Vice versa, we could hope that these new computation techniques might allow us to achieve better understanding of the evolutionary systems that inspired them.
My computer simulations of evolution of the gene networks governing the morphogenesis of early embryos shows the possibility for self-organisation or self-assemblage (or "outgrowth") of gene networks during evolution. This self-assemblage proceeds by means of recruiting of a new gene via closing up of new cascades of interactions between new and old members of the network. (These new genes could appear by way of duplication of the members of this net or another one.) The recruiting of new "netters" does not need to be forced by selection; stabilising selection is quite enough for this. However, if these newly formed gene systems prove to be good for raising the morphological or functional level of organisation, then these Goldschmidt's "hopeful monsters" (Wallace, 1985) can be caught up by driving selection.
After briefly discussing self-organisation in morphogenesis and evolution (in Life and ALife = "artificial life") in Section 2, I'll briefly overview the biological background in Section 3. In Section 4, I'll highlight the idea of gene net growth by recruiting. In Section 5, I'll describe an initial genotype to phenotype transformation scheme ("wild-type" species) and its variability, while in Section 6, I'll show how the model simulates an oversimplified scheme of the gene net appearance. Finally, in Section 7, I will discuss macroevolutionary questions evoked by my computational results.
Get your own Free Homepage 