Evolving a Gene Regulatory Network

From Master Projects
Jump to: navigation, search

has title::Evolving a Gene Regulatory Network
status: finished
Master: project within::Computational Intelligence and Selforganisation
Student name: student name::Danielle de Man
Start start date:=2012/01/01
End end date:=2013/01/31
Supervisor: Evert Haasdijk
Second supervisor: Nicola Bonzanni
Second reader: has second reader::Anton Feenstra
Thesis: has thesis::Media:Thesis.pdf
Poster: has poster::Media:Posternaam.pdf

Signature supervisor



Cellular processes are controlled by networks of interacting genes and molecules. In the cell the product of the expression of one gene can affect the expression of another, creating a network structure in which genes regulate each other through their products. The emergent properties of these and other networks determine the behavior of the cell. Much is unknown about these complex gene regulatory networks. Understanding how these networks work and what they look like provides us with a completer picture of the inner workings of the cell.

For this project we will be focusing on improving the Hematopoietic gene regulatory network. This particular network is responsible for the differentiation of stem cells to a diverse range of blood cells. At each stage of this differentiation the cell stabilises to a distinct steady-state. The emergence of these steady-states are an intrinsic property of the network itself and can be examined by creating a state space graph. In the state space graph each state of the network is represented as a node in the graph. These states are connected in the state space when it is possible to get from one state to the next through a single change in the gene regulatory network. The structure of the state space reveals information about the amount of steady-states the gene regulatory network contains and which genes participate in them. The current Hematopoietic gene regulatory network does not yet concur with measurements made of the cell in each steady-state. We will attempt to improve the network using a multi-objective genetic algorithm in which the fitness function uses the steady-state information as a quality measure of the network. The algorithm will first be employed on a synthetic network, after which it will be used to improve the Hematopoietic network. The final results for the real-life network are tested in a lab.