Date of Award
Doctor of Philosophy (PhD)
The manual process of comparing biological recordings from electrophysiological experiments to their mathematical models is time-consuming and subjective. To address this problem, we have created a blended system that allows for objective, high-throughput, and computationally inexpensive comparisons of biological and mathematical models by developing a quantitative measure of likeness (error function). Voltage recordings from biological neurons, mathematically simulated voltage times series, and their transformations are inputted into the error function. These transformations and measurements are the action potential (AP) frequency, voltage moving average, voltage envelopes, and the probability of post-synaptic channels being open. The previously recorded biological voltage times series are first, translated into mathematical data to input into mathematical neurons, creating what we call a blended system. Using the sea slug Melibe Leonina's swimming central pattern generator (CPG) as our circuit to compare and the source of our biological recordings, we performed a grid search of the conductance of the inhibitory and excitatory synapse found that a weighted sum of simple functions is required for a comprehensive view of a neuron's rhythmic behavior. The blended system was also shown to be able to act as rhythm directors like pacemakers and drivers of Dendronotus Iris swimming interneuron (Si) cells and was able to replicate the perturbations of biological recordings. After verification steps using different configurations, calculated mean and variance of rhythmic characteristics, as well as recordings created from data augmentation. The form of data augmentation introduced can be generalized to other biological recordings or any time series. With all these tools developed and expanding the parameter dimensions a hypothesis was posited that there is a contralateral electric synapse not previously included in the Melibe CPG model.
Bourahmah, Jassem N., "Biological Neuron Voltage Recordings, Driving and Fitting Mathematical Neuronal Models." Dissertation, Georgia State University, 2023.
File Upload Confirmation