PhD proposal: A biologically plausible computer model of pathological neuronal oscillations observed in Parkinson’s disease
PhD supervisor Dominique Martinez (firstname.lastname@example.org)
Co-supervisor Laure Buhry (email@example.com)
Keywords : Computational neuroscience, Parkinson
Qualifications: The candidate should be proficient in one or several of the following areas: applied mathematics, computer science, physics, computational neuroscience or other disciplines related to biophysical modelling. English speaking students are welcome to apply. Applications should include a CV and a letter of motivation and should be sent to Dominique.Martinez@loria.fr and firstname.lastname@example.org no later than May 10, 2018.
Parkinson’s disease is the second most common neurodegenerative disease worldwide with more than one million cases in Europe. The annual cost of its treatment has reached 14 billion euros, with a significant increase expected because of the aging population. The disease impacts the economy, health care systems and the patients’ social well-being. Parkinson’s patients have severe motor problems associated with pathological neuronal synchronization in a particular area of the brain (basal ganglia). Treatment of the symptoms of Parkinson’s disease is possible by deep brain stimulation, i.e. by high frequency electrical stimulation using an electrode implanted in the basal ganglia. It is estimated that 400 new patients every year in France benefit from this therapeutic approach. Yet, the current method has significant side effects because it applies a continuous level of electrical stimulation, regardless of changes in the patient’s needs during the day. Each patient’s brain is dynamic and it is vital to find a personalized approach that combat the symptoms in real time. Therefore, an effective treatment of Parkinson’s disease should be based on a so-called “smart” stimulator where the system first detects and analyzes brain activity and then delivers specific stimulation patterns to each patient according to his/her needs at any given time. To develop this type of closed-loop approach, researchers often resort to animal experiments using rats or monkeys that have been given a drug that removes dopamine from their brain
The thesis works have two objectives. The first is to provide researchers with a biologically plausible model of basal ganglia and a simulation software to test certain hypotheses without first resorting to animal experiments. Existing experimental data collected by our biologist collaborators will be integrated at two levels: individual neurons and interconnected network. For individual neurons, we will apply the general principles of Hodgkin-Huxley type neural models. For the network, the programming of the model will be done in C language by integrating numerically (Runge-Kutta) a system of coupled ordinary differential equations. Due to the large number of neurons involved (of the order of 100,000 neurons in the rat), a parallel implementation (Grig5000) may be necessary. The second objective is to exploit the model to (i) test certain hypotheses on the origin of pathological synchronization observed in Parkinson’s disease and (ii) develop new methods, pharmacological or by closed loop deep brain stimulation, to combat pathological synchronization and motor disorders. The hypotheses we intend to study concern the role of synaptic connections (in particular GABAergic) and intrinsic neuronal properties (in particular SK channels that control the precision of neuronal discharge).