Self-organizing neural architecture and spiking neural model of the tadpole spinal cord
-
1
University of Plymouth, School of Computing and Mathematics, United Kingdom
Traditional approach to a design of neural network model in computational neuroscience includes specification of the model's element, description of the architecture of connections between neurons, and formulation of the learning rule for modification of the connection strength. There is a broad range of biologically realistic models of a single neuron such as the conductance based Hodgkin-Huxley type models and their simplifications, e.g. Morris-Lecar model. As a rule in neuronal modelling, the architecture of neural connections is unknown. A typical approach is to use some standard connection scheme: all-to-all connections, local connections, random connections, feed-forward connections, etc. A more promising approach is to learn from a biological system how neurons self-organise their connectivity pattern to provide a proper functioning of the neural circuit.
We use this approach in our collaborative project with Prof. Alan Roberts Lab, University of Bristol to study a neural network of spinal cord of young Xenopus tadpole (Li et al., 2007, Borisyuk et al., 2008). The spinal cord of tadpole is simple enough to allow detailed neurobiological experimentation with anatomy and neurophysiology of neurons. The spinal cord contains about 2000 neurons which can be arranged into six different cell types relevant to the control of swimming.
Experimental data for this modelling have been provided by A. Roberts. In his Lab there is a unique collection of experimental data which includes both anatomical measurements on distribution of cell bodies, dendrites, axons in the spinal cord cylinder and multiple experimental data on electrophysiology of different neurons (Li et al., 2007).
To define the connection architecture, we model a process of neural network development. A simple model of axon growth is a basis for the self-organisation process which results in the complete biologically realistic architecture of connections. Figure shows a fragment of reconstruction of the tadpole spinal cord. Each cell type is represented by a separate colour: axons, dendrites, initial point of axons, and synapses are shown. The dendrite is represented by a vertical bar and the synapse is shown by a circle. A colour of the circle relates to presynaptic neuron. More details are given in Borisyuk et al., 2008. The model includes both deterministic and stochastic components and allows to generate multiple copies of the spinal cord architecture with the same statistical characteristics of axon distribution as the measurements of real axons.
We combine the generated anatomy with Morris-Lecar model of spiking neurons to find whether this neural network can produce a specific pattern of neural activity corresponding to the swimming of tadpole. Preliminary simulations show that the neural network with randomly distributed parameter values can swim.
References
1. Borisyuk R., Cooke T., Roberts A. (2008) Stochasticity and functionality of neural systems: Mathematical modelling of axon growth in the spinal cord of tadpole. BioSystems, 93:101-114
2. Li W.-C., Cooke T., Sautois B., Soffe S., Borisyuk R. and Roberts A. (2007) Axon and dendrite geography predict the specificity of synaptic connections in a functioning spinal cord network. Neural Development, 2:17
Conference:
Neuroinformatics 2009, Pilsen, Czechia, 6 Sep - 8 Sep, 2009.
Presentation Type:
Poster Presentation
Topic:
Computational neuroscience
Citation:
Borisyuk
R
(2019). Self-organizing neural architecture and spiking neural model of the tadpole spinal cord.
Front. Neuroinform.
Conference Abstract:
Neuroinformatics 2009.
doi: 10.3389/conf.neuro.11.2009.08.100
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
25 May 2009;
Published Online:
09 May 2019.
*
Correspondence:
Roman Borisyuk, University of Plymouth, School of Computing and Mathematics, Plymouth, United Kingdom, rborisyuk@plymouth.ac.uk