Generalized contexts for reaction systems: definition and study of dynamic causalities

Reaction systems are a qualitative formalism for the modelling of systems of biochemical reactions. In their original formulation, a reaction system executes in an environment (or context) that can supply it with new objects at each evolution step. The context drives the behaviour of a reaction system: it can provide different inputs to the system that can lead to different behaviours. In order to more faithfully deal with open systems, in this paper we propose a more powerful notion of context having not only the capability to provide objects, but also to absorb (or remove) objects at each evolution step. For such reaction systems with generalized context we investigate properties of dynamic causality by revising the previously proposed concept of formula based predictor. A formula based predictor is a Boolean formula characterising all contexts that lead to the production of a certain object after a given number of steps. In this paper, we revise the theory of formula based predictors in order to deal with reaction systems executed in a context of the new kind. As applications, we show an example of interaction between biochemical pathways and a reaction system modelling cell metabolism and respiration.