Doctoral thesis (Dissertations and theses)
Detection and characterization of critical transitions in mitochondrial activity via high content screening
Trefois, Christophe


Full Text
Author postprint (74.35 MB)
Request a copy

This work is published under CC BY-NC-ND

All documents in ORBilu are protected by a user license.

Send to


Keywords :
High-content imaging; Critical transitions; Tipping points; Criticality; Alternative Stable States; HeLa; Microscopy; High-content screening; HCS; Parkinson's disease
Abstract :
[en] Critical transitions exist in many dynamical systems, ranging from the Earth’s cli- mate system to microcosm populations. During a critical transition, the state of a dynamical system abruptly changes from one stable state to another, typically without obvious prior warning. Preventing such abrupt changes remains a chal- lenge, however recently, several metrics were suggested as early warning signals. These indicators are thought to have predictive value for upcoming critical transi- tions. In Parkinson’s disease, there are no detectable motor symptoms in a patient until neuronal dopaminergic cell death exceeds 60–70%. Being able to define early warning signals in a disease context could open new avenues for both preventive and disease modifying treatments. We hypothesize that the dynamics of progression of some disorders including Parkinson’s disease could be manifested by critical tran- sitions. However, before rushing into medical applications, a thorough framework needs to be developed that aims to describe such nonlinear dynamics in cellular systems. In this thesis, we set out to study critical transitions in a simple cellular model using mitochondrial membrane potential ∆Ψ m as readout. To identify criti- cal transitions, we established a modular high-content screening platform allowing systematic perturbation of oxidative phosphorylation. To increase the probability for detecting a critical transition in ∆Ψ m , five inhibitory compounds were combined in multiple pairwise concentration landscapes. We show that critical transitions, de- tectable via ∆Ψ m , are an intrinsic property of the cellular system studied and that two-component Gaussian mixture models adequately capture the dynamics of the critical transition occurring for the combination of Oligomycin A and Antimycin A. Adding to that, we identified the coefficient of variation as a strong early warning signal for the upcoming of the critical transitions. This thesis should serve as a foundation for a broader application of critical transitions and early warning sig- nals in both cell culture systems and translational studies aiming to understand the nonlinear dynamics of biological systems.
Research center :
Luxembourg Centre for Systems Biomedicine (LCSB): Experimental Neurobiology (Balling Group)
ULHPC - University of Luxembourg: High Performance Computing
Disciplines :
Life sciences: Multidisciplinary, general & others
Author, co-author :
Trefois, Christophe ;  University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB)
Language :
Title :
Detection and characterization of critical transitions in mitochondrial activity via high content screening
Defense date :
11 August 2014
Number of pages :
Institution :
Unilu - University of Luxembourg, Luxembourg, Luxembourg
Degree :
Docteur en Biologie
Promotor :
President :
Jury member :
Bouvry, Pascal 
Gunzer, Matthias
Walter, Thomas
Name of the research project :
Funders :
FNR - Fonds National de la Recherche [LU]
Available on ORBilu :
since 24 August 2014


Number of views
160 (33 by Unilu)
Number of downloads
8 (7 by Unilu)


Similar publications

Contact ORBilu