This title appears in the Scientific Report :
2023
Multi-Scale Spiking Network Model of Human Cerebral Cortex
Multi-Scale Spiking Network Model of Human Cerebral Cortex
Background: The structure of the brain plays a crucial role in shaping its activity. However, the link between structural connectivity and observed neuronal activity remains incompletely understood. Previous research utilizing a large-scale spiking network model of leaky integrate-and-fire neurons h...
Saved in:
Personal Name(s): | Pronold, Jari |
---|---|
Meegen, Alexander van / Vollenbröker, Hannah / Shimoura, Renan (Corresponding author) / Senden, Mario / Hilgetag, Claus C. / Bakker, Rembrandt / van Albada, Sacha | |
Contributing Institute: |
Computational and Systems Neuroscience; INM-6 Jara-Institut Brain structure-function relationships; INM-10 Computational and Systems Neuroscience; IAS-6 |
Imprint: |
2023
|
Conference: | 2nd Cologne Neuroscience Day, Cologne (Germany), 2023-10-26 - 2023-10-26 |
Document Type: |
Poster |
Research Program: |
Brain-Scale Simulations Human Brain Project Specific Grant Agreement 3 Heterogenität von Zytoarchitektur, Chemoarchitektur und Konnektivität in einem großskaligen Computermodell der menschlichen Großhirnrinde SPP 2041: Computational Connectomics Neuroscientific Foundations |
Publikationsportal JuSER |
Background: The structure of the brain plays a crucial role in shaping its activity. However, the link between structural connectivity and observed neuronal activity remains incompletely understood. Previous research utilizing a large-scale spiking network model of leaky integrate-and-fire neurons has addressed this question for macaque cortex [1,2]. Here, a similar framework is employed to investigate human cortex in a model that links the cortical network structure to the resting-state activity of neurons, populations, layers, and areas.Objectives: The objective of this study is to investigate the link between structural connectivity and observed neuronal activity in human cortex using a large-scale spiking network model, and to create a platform for multi-scale in silico studies of human cortex.Materials and Methods: The model includes all 34 areas in a single hemisphere of human cortex according to the Desikan-Killiany parcellation. Our approach integrates cortical data on architecture, morphology, and connectivity into a multi-scale framework for predicting neuron connections. Each cortical area is represented by a 1 $mm^2$ layered microcircuit adapted from [3] with the full density of neurons and synapses. Inter-area connectivity relies on diffusion tensor imaging data [4] and the determination of laminar patterns of synaptic connectivity takes into account human neuron morphology data [5]. The model comprises 4 million neurons and 50 billion synapses, simulated with the NEST simulator on the supercomputer JURECA-DC. Results and Conclusions: Simulations of the model with uniform synaptic weights reveal a state with asynchronous and irregular activity that deviates from experimental recordings in terms of spiking activity and inter-area functional connectivity. Increasing inter-area synapse strength enables the model to capture both microscopic and macroscopic resting-state activity of human cortex measured via electrophysiological recordings and fMRI [6]. Furthermore, the model reveals rapid propagation of the effects of a single-spike perturbation across the entire network. This suggests individual spikes play a role in fast sensory processing and behavioral responses in the cortical network. Overall, the model serves as a basis for the investigation of multi-scale structure-dynamics relationships in human cortex. |