Modelling and analysis of large-scale models of signalling networks

  • Modellierung und Analyse von Großmodellen von Signalisierungsnetzen

Gjerga, Enio; Mitsos, Alexander (Thesis advisor); Saez-Rodriguez, Julio (Thesis advisor)

Aachen : RWTH Aachen University (2020, 2021)
Dissertation / PhD Thesis

Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2020


Cells rely on a system of signal processing and transmission networks to perform and coordinate their basic activities. The transmission process is governed through chemical signals and reactions which can be mediated by proteins or other smaller molecules. This process is collectively named cell signalling. Errors in cell signalling can lead to the development of severe diseases such as cancer. As such, an understanding at a system level of cell signalling can help us identify aberrant processes which can be therapeutically targeted. Considering the huge variety of proteins and their modifications, these kinds of systems can be quite large, complex and dynamic. This makes the modelling of cell signalling systems particularly challenging. The aim of the PhD project consists of developing new and existing methods used for the modelling cell signalling networks for the understanding of processes involved in it and the mechanisms of cell function. This was achieved through the development of various modelling tools (CARNIVAL, PHONEMeS, CellNOptR and Dynamic-Feeder). At the core of these tools stands the integration of optimisation techniques with simulation analysis in order to provide information about the behaviour of the biological system representing our signalling networks. Such an approach allows the prediction of perturbation outcomes (i.e. treating cells with drugs) and can give help in the planning of prospective treatment strategies. Model-based design of cell signalling systems, especially optimisation-based, is a major focus of this work, and here I will be mostly focusing on understanding, exploiting and designing methods of integer linear and non-linear dynamic optimisation techniques for structural and parametric identification of large signalling networks based on experimental data. The approaches followed, allow for a more efficient analysis of large-scale models of signalling networks and ever-increasing data.