In silico experimental evolution shows that complexity can rise even in simple environments

Guillaume Beslon 1, 2 Vincent Liard 1, 2 David P. Parsons 1, 3 Jonathan Rouzaud-Cornabas 1, 2
1 BEAGLE - Artificial Evolution and Computational Biology
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information, Inria Grenoble - Rhône-Alpes, LBBE - Laboratoire de Biométrie et Biologie Evolutive, CarMeN - Cardiovasculaire, métabolisme, diabétologie et nutrition
Abstract : Systems biology is often viewed as reverse engineering of biological systems. However, contrary to reverse engineering, systems biology deals with objects that have not been designed, that have no given purpose and that don’t follow engineering rules (e.g. modularity, standardization…). Indeed, we don’t know what are the “design rules” that evolution imposes to biological systems while this knowledge would be a valuable interpretative framework for systems biology. One of the recurrent questions on that matter is the origin of the striking molecular complexity of biological systems. Answering this question requires deciphering the complex interactions between all the forces that drive evolution, including selective and non-selective ones. In this context, simulation is a valuable tool as it enables to observe how organisms grow in complexity (or not) when they evolve in environments which complexity is perfectly mastered. In Liard et al., 2018, we used the Aevol platform (, to design an in silico experiment in which populations of organisms evolved in an environment designed to enable survival of the simplest possible organism (i.e., an organism whose genome encodes a single gene) and in which this simple organism have the best possible fitness. By repeatedly evolving organisms in this experimental design, we observed two very different outcomes: some lineages were able to quickly find the optimal genotype (one single gene) and were then stable for the rest of the experiment. However, most lineages were not able to find the optimal genotype and showed a very different dynamics with continuous complexification through gene acquisition all along the experiment. Importantly, these “complex” organisms ended up with fitness values typically 10 to 100 times lower than the simple ones. Our results show that, in such a simple constant environment, there is a decoupling between the molecular complexity of the organisms and the complexity of the environment. This shows that selection for complexity is not mandatory for complexity to evolve and that complex biological structures could flourish in conditions where complexity is not needed. Reciprocally, the global function of complex biological structures could very well be simple. We think this result is greatly significant for both evolutionary biology and systems biology. Liard, et al. (2018) The Complexity Ratchet: Stronger than selection, weaker than robustness. In: Proceedings of ALife 2018.
Type de document :
ICSB 2018 - 19th International Conference on Systems Biology, Oct 2018, Lyon, France. pp.1
Liste complète des métadonnées
Contributeur : Vincent Liard <>
Soumis le : jeudi 29 novembre 2018 - 13:52:26
Dernière modification le : mardi 4 décembre 2018 - 10:33:04


Poster ICSB18.pdf
Fichiers produits par l'(les) auteur(s)


  • HAL Id : hal-01939365, version 1



Guillaume Beslon, Vincent Liard, David P. Parsons, Jonathan Rouzaud-Cornabas. In silico experimental evolution shows that complexity can rise even in simple environments. ICSB 2018 - 19th International Conference on Systems Biology, Oct 2018, Lyon, France. pp.1. 〈hal-01939365〉



Consultations de la notice


Téléchargements de fichiers