EN FR
EN FR


Section: New Results

Introduction

Participant : Andreea Beica.

The PhD of Andreea Beica [9] aims at studying two aspects related to the modelling of Biochemical Reaction Net- works, in the context of Systems Biology.

In the first part, we analyse how scale-separation in biological systems can be exploited for model reduction. We first argue for the use of rule-based models for prototyping genetic circuits, and then show how the inherent multi-scaleness of such systems can be used to devise a general model approximation method for rule-based models of genetic regulatory networks. The reduction proceeds via static analysis of the rule system. Our method relies on solid physical justifications, however not unlike other scale-separation reduction techniques, it lacks precise methods for quantifying the approximation error, while avoiding to solve the original model. Consequently, we next propose an approximation method for deterministic models of biochemical networks, in which reduction guarantees represent the major requirement. This second method combines abstraction and numerical approximation, and aims at providing a better understanding of model reduction methods that are based on time- and concentration- scale separation.

In the second part of the thesis, we introduce a new re-parametrisation technique for differential equation models of biochemical networks, in order to study the effect of intracellular resource storage strategies on growth, in self-replicating mechanistic models. Finally, we aim towards the characterisation of cellular growth as an emergent property of a novel Petri Net model semantics of Biochemical Reaction Networks.