Methods and algorithms for the study of biomolecular networks
Doctoral thesis
Permanent lenke
http://hdl.handle.net/11250/2641874Utgivelsesdato
2019Metadata
Vis full innførselSamlinger
Sammendrag
Living organisms play a crucial role in our lives, serving us as sources of food and materials. In order to effectively work with living organisms, we need to understand them. And we understand biological organisms by modelling them. This work is centered on several topics of biological modelling, with the perspective ranging from single molecule collisions to microorganism proliferation. We also develop algorithms to correct existing models and make new ones.
In the first part of this work we focus on the modelling of systems where the movements of individual molecules affect the fate of the organism as a whole. This is called stochastic modelling, which we apply to understand the inner workings and control of genetic switches, which can be used in for instance in chemical production.
In the following parts, we change perspective to a broader one, focusing on the modelling of whole organisms, especially using a method called Flux Balance Analysis. We discuss the present day approaches to modelling and propose new and better ones. In this process we discover that most present day models suffer from large numbers of errors, called inconsistencies. We therefore devise algorithms that would find and show us what the causes of these errors are.
Finally we discuss how we can make "model models" of living organisms, using artificial evolution instead of experimental data. In this process we discover that mimicking natural selection in natural environments may be a better approach to generating life-like artificial models than trying to change model properties in the way we think they should be.
All of these insights should aid future modellers in making correct and realistic models, empowering the progress of the field of biological modelling and biotechnology in general.