Modulating Neuron Behavior

Full Title

Implementation of Software and Neuromorphic Hardware to Modulate Neuron Behavior

Marketing Title

Modulating Neuron Behavior

Acknowledgements

Special thanks to our mentors Dr. Frederic Broccard (fbroccard@ucsd.edu), Dr. Eduardo Macagno (emacagno@ucsd.edu), and Soumil Jain (s3jain@ucsd.edu) for guiding us, to Dr. Gert Cauwenberghs for allowing us to use his laboratory, and to Dr. Bruce Wheeler and Sicily Rose Panattil for helping us in Senior Design.

Abstract

Half center oscillators (HCOs) are inhibitory coupled neuron pairs that can be connected to other neurons to form rhythmic patterns in the body. We simulated two HCOs when they are joined with excitatory feed-forward and feed-back connections using a brute-force approach in MATLAB to understand their parameter space under the Izhikevich and Morris-Lecar neuron models. Additionally, we built a dynamic conductance clamp to demonstrate that adding a shunt and a sodium conductance modulates a leech ganglion Retzius cell. The added shunt conductance led to the cells increasing their spiking frequency. Adding a sodium conductance between 100 and 140nS creates a bursting pattern of action potentials. Future researchers can now use the dynamic clamp and its capabilities to model HCOs in a leech ganglion and compare the in vitro outcomes to the simulation results.

Problem Statement

The absence of a simulation database for coupled half center oscillators (HCOs) hinders the exploration of the parameter space and implementation of the model in hardware. Since there is not a database of coupled HCOs, researchers do not have the access to data they need to fully explore the parameter space beyond biological constraints nor the reliability of a non-organic system. The collection of cellular voltages dependent on the many adjacency matrices that this project is producing will allow researchers to peruse the theoretical voltages quickly and at leisure without having to set up complex biological experiments.

The combination of a dynamic conductance clamp with real leech neurons will allow researchers to perform a wider variety of experiments and be able to change variables in a more precise manner. For example, they could replicate the effects of drugs without actually using them.

Project Video