Artificial life VII: looking backward, looking forward (editor's introduction to the special issue).

Abstract

This issue contains highlights from the Artificial Life VII conference, held at Reed College in August 2000 [1]. A few authors who presented papers at the conference were invited to revise and expand their papers for publication in this special issue. Freed from the artificial space limits of conference proceedings, their papers provide additional details and situate their work in a broader context. One additional paper on open problems in artificial life grew out of discussion at the end of the conference. All the papers show the breadth and depth of the work presented at the conference. All address very deep questions about living systems, and most build bridges to concrete biological data and generate experimentally testable predictions. The origin of multicellularity is one of the major transitions in the evolution of life, and differentiated multicellular life has apparently evolved independently many times. The developmental processes which create these multicellular organizations share three notable features. First, development starts with a homogeneous cell (or set of cells) that are multipotent, that is, have the ability to differentiate into many different types of specialized cells. Second, the developmental process has an intrinsic arrow of time because the differentiated cells that result are not multipotent. Third, the developmental process is stable in the face of perturbations that destroy clusters of cells. In “Complex Organization in Multicellularity as a Necessity in Evolution,” Chikara Furusawa and Kunihiko Kaneko provide a minimal model of cellular dynamics that explains these universal features of multicellular differentiation. Furusawa and Kaneko numerically simulated a one-dimensional chain of cells governed by randomly generated biochemical reaction networks. They found that, while competition for resources prevents resources from flowing through chains of undifferentiated cells, cellular differentiation allows cells to share resources throughout the cellular chain. Thus, although individual cells taken from undifferentiated chains can grow more quickly than cells from differentiated chains, chains of differentiated cells can grow more quickly than chains of undifferentiated cells. These results lead to a number of experimentally testable predictions about multicellularity in biological organisms and provide a mechanism by which multicellularity inevitably arises in the evolution of cellular differentiation. The genetic code is a nearly universal feature of life on earth, although it does have a few exceptions such as mitochondrial DNA. It is difficult to understand at first how the genetic code could evolve at all; a mutation at a given codon might be a selective advantage, of course, but any change in the code would entail wholesale changes at a vast proportion of codons. How could this but be disastrous? Hiroaki Takagi, Kunihiko Kaneko, and Tetsuya Yomo propose a solution to this problem in