Using mathematics to disentangle the role of synaptic communication

点击翻译按钮获取中文摘要

This fellowship will provide new understanding on how neuronal networks encode and store information by incorporating mathematical models in electrophysiology experiments to enable direct manipulation, at cellular scales, of synaptic communication parameters including synaptic conductances, axonal and propagation delays, and network connectivity features. Neuronal networks operate through a combination of spiking electrical activity generated in individual neurons and communication (typically through synapses) between cells, which together allow the network to generate a variety of distinct patterns of electrical activity. In turn, these electrical activity patterns underpin a wide range of neuronal computation in sensing, learning and memory.Decades of electrophysiological research, supported by mathematical modelling, have yielded key insights into the response properties of individual neurons to stimuli. Far less progress has been made in understanding how communication properties shape electrical network rhythms. In part, this is due to the difficulty in ascertaining the so-called 'wiring diagram' that details which neurons communicate with one another. In addition, there are no experimental tools to directly modify communication properties at fine scales. Were such tools available, they would facilitate quantitative investigations that directly link communication properties to network rhythms. This fellowship will develop such tools by embedding mathematical models using closed-loop real-time feedback during electrophysiological recordings to enable direct manipulation of communication parameters.The tools developed during this fellowship will be used to characterise and quantify the role of synaptic parameters, such as synaptic conductances and transmission delay, in the generation of electrical rhythms in neuronal networks. I will first develop a mathematical model of synaptic communication in a cultured neuronal network. This model, which will be calibrated to patch clamp and voltage imaging recordings of network activity, will account for the dynamics of synaptic communication as well as the network wiring diagram. The model will be analysed through bifurcation analysis and numerical simulation to predict how the number and type of electrical rhythms supported by the network changes with respect to:1) variation of communication parameters such as synapse conductance and axonal and dendritic propagation delays; 2) heterogeneity in communication parameters across the network;3) synaptic plasticity, in which synaptic conductances vary in response to network activity. A closed-loop control strategy will be used to test these predictions by modulating the synaptic communication properties in real neuronal networks in the same way as in the mathematical model. The fellowship will thus provide a framework for hypothesis generation and testing on the contribution of individual synaptic parameters to neuronal network rhythms. The first phase of the fellowship (years 1-4) will use cultured cell lines to minimise its ethical costs. In the second phase (years 5-7), studies will be performed in neuronal networks cultured from patient-derived induced pluripotent stem cells, made available through collaboration with experimental partners, to investigate how synaptic deficits contribute to the aberrant electrical rhythms observed in these networks.This fellowship will provide deeper understanding of the fundamental mechanisms associated with neuronal network functioning at small scales. This improved understanding may help, in the long term, to treat disorders such as autism and motor neuron disease. In the shorter term, this will accelerate the development of so-called biological neuronal networks that attempt to harness the complexity of cultured neuronal networks to improve the performance of machine learning algorithms by replacing computer chips with real neurons.