PhD project: Multiscale Models to Link Cellular Mechanisms of Dopamine Neuromodulation to whole-brain Dynamics

Neuromodulation is characterized by changes in the biophysical properties of single neurons that orchestrate shifts in whole-brain activity and function [1] not solely at individual neurons’ level but through the complex dynamics of mesoscopic neural ensembles. The comprehension of this multiscale mapping stands as a pivotal endeavor, especially concerning diseases such as Parkinson’s and psychiatric disorders, where it holds significant implications. 

Quantitative models that bridge microscale neuronal neuromodulation with systems-level brain function underscore gaps in knowledge and offer avenues for integrating theoretical and experimental work. Efforts to develop models accounting for population-level activity modulated by single-neuron properties rely on key methods such as mean-field reduction. This approach describes the state of a large population using the first moments of the population variable’s distributions. 

At the Institut de Neurosciences de Systemes (INS), this methodology has been applied [2] to theoretically elucidate the effects of the extracellular potassium concentration at the population activity emerging from ionic mechanisms that operate at the single-cell level [3]. It offers a mathematical framework for analyzing large-scale neuronal activity, focusing on aggregate properties rather than individual neuron dynamics. 

Similarly, our PhD project aims to develop a theoretical framework integrating large-scale modeling with biophysical detail, focusing on distilling predictive data from biophysically detailed models to capture essential features relevant to neuromodulation effects. The initial phase involves constructing a data-driven biophysical model at the cellular level, incorporating dynamics related to neuromodulation, particularly variations in local dopamine concentration. Subsequently, this single-cell model will inform large neural network simulations, followed by mean-field reduction to a neural mass model. This dimensionally reduced neural mass model will serve as a foundation for building whole-brain models that can be validated with clinical data.  

At the end of this project, we expect to have developed and characterized the dynamics of multiscale models accounting for the neuromodulation of localized dopamine at the cellular, population, and whole-brain levels. By deepening our understanding of neuromodulatory dynamics, these models play an essential role in understanding neurological conditions. These models hold the potential to be used for personalized brain model as it has been previously developed at the INS for epilepsy [4], [5] and advance nosology refinement, facilitating early detection and accurate personalized treatments of neurological disorders such as Parkinson’s disease.  

Under Viktor Jirsa’s supervision and Damien Depannemaecker’s co-supervision at INS, the project benefits from extensive experience in multiscale modeling, experimental data collection, and collaborative efforts, ensuring its feasibility. 

The ideal candidate would possess an interdisciplinary background in physics with programming experience and a keen interest in neuroscience or a neuroscience background with strong programming and mathematical analysis skills. Prior experience with cellular biophysical mechanisms, single-cell modeling, spiking neural networks, dynamical systems, and derivation of mean-field models would be advantageous. 

Keywords: dopamine neuromodulation, multiscale modeling, computational neuroscience, computational psychiatric, whole-brain models. personalized medicine 

Project workflow overview. Data-driven model building and scale integration: the models are constrained by the data at the cellular level and whole-brain level. The models integrate over scale thanks to methods such as mean-field derivation, once built it enables the study of the effect at the whole-brain level of changes at the local microscopic level. Thus, this modeling framework brings an understanding of how local actions through stimulation or pharmacological intervention lead to dynamical changes noticeable at the whole-brain level. The strength of this approach is the ability to keep the biophysical interpretation of biophysical parameters over the scales. 

[1] Hansen, J.Y., Shafiei, G., Markello, R.D. et al. Mapping neurotransmitter systems to the structural and functional organization of the human neocortex. Nat Neurosci 25, 1569–1581 (2022). https://doi.org/10.1038/s41593-022-01186-3 

[2] Damien Depannemaecker, Anton Ivanov, et al.. A unified physiological framework of transitions between seizures, sustained ictal activity and depolarization block at the single neuron level. Journal of Computational Neuroscience, 50(1):33–49, January 2022. ISSN 1573-6873. doi 10.1007/s10827-022-00811-1.  

[3] Abhirup Bandyopadhyay, Giovanni Rabuffo, et al. Mean-field approximation of network of biophysical neurons driven by conductance-based ion exchange. Nov. 2021. doi: 10.1101/2021.10.29.466427.  

[4] Huifang E. Wang, Marmaduke Woodman, et al. Delineating epileptogenic networks using brain imaging data and personalized modeling in drug-resistant epilepsy. Science Translational Medicine, 15, 1 2023. ISSN 1946-6234. doi: 10.1126/scitranslmed.abp8982.  

[5] Viktor Jirsa, Huifang Wang, et al. Personalised virtual brain models in epilepsy. The Lancet Neurology, 3 2023. ISSN 14744422. doi:10.1016/S1474-4422(23)00008-X.