Ph. D. Project
Title:
BIOLOGICALLY REALISTIC COMPUTATIONAL SIMULATION OF THE EPILEPTIQUE HIPPOCAMPUS, TOWARDS THERAPEUTIC TARGETS
Dates:
2021/10/01 - 2024/09/30
Supervisor(s): 
Other supervisor(s):
BUHRY Laure (lbuhry@loria.fr)
Description:
The thesis continues the PhD work by Amélie Aussel under the supervision of the same supervisors, Laure Buhry and Radu Ranta.
We would like to extend the proposed model to incorporate, in the context of mesial temporal lobe epilepsies, other factors that
could influence the generation and the measurement of pathological hippocampal activity. In addition to understanding how this
structure works, the objective is to include electrical stimulation and pharmacological treatments.

The student will have access to surface and depth cerebral data (measurements and positions of the EEG / SEEG
electroEncephaloGraphy and Stereo-ElectroEncephaloGraphy electrodes) as well as, if possible, to micro-scale measurements
(neuronal action potentials and LFP, Local Field Potential, simultaneously). These signals, recorded in patients with mesian
temporal lobe epilepsies, will always serve as basic elements for the construction of models and validation tools.

Several research directions are being considered:

1) enrich the structure of the model by incorporating different types of interneurons present in the hippocampus. We predict that
this will be of decisive importance in the dynamics and resonance of recruited neural networks because interneurons exhibit
frequencies of emission of preferential action potentials from one type to another (Hartwich et al. 2009, Komendentov et al. al.
2019, Pelkey ​​et al. 2017). In addition, certain types of interneurons have a greater sensitivity at the level of their extra-synaptic
receptors (ie outside the places commonly observed in communication between neurons), which can be critical in the
development of new therapeutic targets because pharmaceutical treatments crossing the blood-brain barrier generally target
synaptic and extra-synaptic receptors without distinction.

2) more finely model the measurement acquired by the different types of electrodes by taking into account the locations of
synaptic projections on the pyramidal neurons. While it is accepted that the latter are the main contributors to LFP (extracellular
electrical potentials measured by the electrodes), it is not clear which are the preponderant synapses (the synapses, excitatory or
inhibitory, can have different projections on the morphology of the postsynaptic pyramidal neuron). [Telenczuk 2017, 2019] In
addition, we will also assess the contribution of action potentials to extracellular measurements [Aussel 2019, Scheffer-Teixeira
2013].

3) adjust and, as far as possible, optimize the new parameters of the proposed model by comparing its outputs to the signals
provided by other models in the literature, less biologically precise but more mathematically tractable, particularly in terms of
connections between populations different structures (intra and inter structure) [Wendling 2010, Ursino 2010, Xiang 2017], which
will also allow the model to be extended to hippocampo-cortical interactions.

4) If time permits, the work of modeling healthy hippocampal activity will be extended to other stages of sleep. To our
knowledge, there is currently no single model capable of reproducing the different hippocampal rhythms due to the variability of
the synaptic time constants involved and the complexity of the interactions between neurotransmitters. The understanding of the
"healthy" system will be useful in the management of the pathological rhythms of the mesian temporal lobe.
Keywords:
computational neuroscience, mathematical modeling, hippocampus, epilepsy, sleep, human
Conditions:
A master 2 in computational neuroscience, computer science, signal processing or applied mathematics is required, as well as a
strong interest in neurosciences although no background in biology is initially required.
Strong programming skills required, in particular in Python. Additional knowledge in C and Matlab is preferred.
Knowledge in parallel computing would be beneficial.
Department(s): 
Biology, Signals and Systems in Cancer and Neuroscience