The BioDyn language and simulator. Application to an immune response and E. coli and phage interaction

Nowadays multiagent systems (MAS for short) [1] are used as a tool to understand the structure, the behaviour and the functions of biological processes. Indeed, MAS enable living organisms to be simulated in order to perceive what characterizes these selforganizing systems. They are used in order to discern how these organisms can produce achievements that exceed the sum of all the performances of their components. A MAS is a system of multiple active objects, called agents, which have the ability to perceive, act, interact and move in their environment. Each of the MAS considered here is constituted by the following components : (i) An environment, which is a three-dimensional space; (ii) One or several categories of agent; two agents, which are not from the same category, differ in the simulator by their colour and/or their shape and/or their behaviour; (iii) A small set of governing agent behaviour is associated to each category. A cell can be divided into numerous autonomous entities interacting with each other: molecules, ions, membranes and organelles. These entities can be seen as interacting reactive agents. In this way, a reactive agent is an abstraction or a simplification of a biological entity. It has sensors to get information from its local environment, a simple behaviour to change its internal state, and actuators to modify the environment. Much data comes from large scale biology. These data mainly concern the genome but they also deal with cell physiology, including membranes, hyperstructures, and compartments. To understand the cell functionalities the genome is essential; nevertheless, the physiological environment of the DNA The BioDyn language and simulator. Application to an immune response and E. coli and phage interaction P. Ballet, A. Zemirline and L. Marcé