Metabolic Pathway Analysis : from small to genome-scale networks
The need for mathematical modelling of biological processes has grown alongside with the achievements in the experimental field leading to the appearance and development of new fields like systems biology. Systems biology aims at generating new knowledge through modelling and integration of experimental data in order to develop a holistic understanding of organisms. In the first part of my PhD thesis, I compare two different levels of abstraction used for computing metabolic pathways, constraint-based and graph theoretical methods. I show that the current representations of metabolism as a simple graph correspond to wrong mathematical descriptions of metabolic pathways. On the other hand, the use of stoichiometric information and convex analysis as modelling framework like in elementary flux mode analysis, allows to correctly predict metabolic pathways. In the second part of the thesis, I present two of the first methods, based on elementary flux mode analysis, that can compute metabolic pathways in such large metabolic networks: the K-shortest EFMs method and the EFMEvolver method. These methods contribute to an enrichment of the mathematical tools available to model cell biology and more precisely, metabolism. The application of these new methods to biotechnological problems is also explored in this part. In the last part of my thesis, I give an overview of recent achievements in metabolic network reconstruction and constraint-based modelling as well as open issues. Moreover, I discuss possible strategies for integrating experimental data with elementary flux mode analysis. Further improvements in elementary flux mode computation on that direction are put forward.
Use and reproduction:
All rights reserved