Evolved Developmental Strategies of Artificial Multicellular Organisms

Abstract

We present the use of a new computationaly efficient 3D physics model for the simulation of cells in a virtual sea world. In this model, cells can freely assemble and disconnect along the simulation without any separation between the developmental and evaluation stages, as is the case in most evo-devo models which only consider one cell cluster. While allowing for the discovery of interesting behaviors through the addition of new degrees of freedom, this 3D center-based physics engine and its associated virtual world also come with their drawbacks when applied to evolutionnary experiments: larger search space and numerous local optima. In this paper, we have designed an experiment in which cells must learn to survive by keeping their genome alive as long as possible in a demanding world. No morphology or strategy is explicitly enforced; the only objective the cells have to optimize is the survival time of the organism. We show that a novelty metric, adapted to our evo-devo matter, improves the outcome of the evolutionary runs. This paper also details some of the developmental strategies the evolved multicellular organisms have found in order to survive.

Artificial Gene Regulatory Networks—A Review

“…Interface genes In 2016, this work was extended to the third dimension and explored the developmental strategies in a more realistic environment in which cells must proliferate both in soil that contains nutritive resources and in the air where sunlight transform the nutriment into energy [40]. Once again, the fitness function only consisted of the survival duration of the organism.…”

Self-sustainability Challenges of Plants Colonization Strategies in Virtual 3D Environments

“…This crossover operator differs from those commonly found in the literature [4,6,15] on three points: (1) it can fail early on, (2) is biased by the female genome and (3) has low resistance to large structural differences. The rationale behind point 3 is that, instead of devising a robust operator that can produce a somewhat viable offspring from two completely unrelated individuals, a minimalist alignment procedure is better suited to sexual reproduction of same species creatures in which the population is mostly homogeneous.…”

“…its synthetic behavioral description, is (F a , F b , F p , R, G, S), with R the number of successful autonomous reproductions, G the number of autonomous generations and S the seed size. In each scenario, plants are evaluated on two to five criteria using a tournament selection where 3 participants are randomly selected from the population and compete on a random objective as described in [6].…”

“…Closer to biology is the use of Genetic Regulatory Networks to control cellular building blocks. By using such a finer-grained representation it was shown possible to evolve cell clusters towards specific shapes [5] or generate creatures with self-organized organs in non-trivial environments [6].…”

Speciation under changing environments

“…Further generalization led to the biologically inspired Genetic Regulatory Networks which, by defining the cell as the elementary unit, emulate its internal chemistry through self-interacting 'proteins' controlling its life-cycle. Using such a low-resolution building block allowed for the generation of specific shapes (Joachimczak and Wróbel, 2008) and organ emergence for creatures embedded in a virtual environment (Disset et al, 2016).…”