File(s) under embargo
until file(s) become available
Statistical Analysis of Calcium-Imaged Population Activity
Calcium imaging has been widely adopted for its ability to record from large populations of labelled cells within the brain. Given that we know neurons work together in any task, we can study how populations of neurons covary with each other. Such statistical studies of population activity have been particularly useful in electrophysiology studies in helping our understanding about neural circuit mechanisms, and these analysis tools are only just beginning to be used in calcium imaging studies.
The focus of this thesis is to provide a foundation for the statistical analysis of calcium imaged recordings of large populations of cells. In the first part of this thesis, we focus on population analysis techniques traditionally used on electrophysiological recordings, and ask the question of whether these techniques are appropriate for calcium imaged recordings of neuronal populations, and how they may be applied. In this process, we also developed a simultaneous dimensionality reduction and deconvolution method, termed Calcium Imaging Linear Dynamical System (CILDS). We found that when performing dimensionality reduction, there is benefit in leveraging population statistics to perform deconvolution for dimensionality reduction, as done in CILDS.
In the second part of this thesis, we evaluated these different dimensionality reduction methods in calcium-imaged recordings from the dorsal raphe nucleus in larval zebrafish and V1 in mouse. We found that despite the different animal models and brain areas, CILDS outperformed other approaches towards dimensionality reduction on calcium-imaged neuronal populations. CILDS was better able to peer through the calcium decay and apply the appropriate degree of temporal smoothing to the latent variables.
With our understanding of population analysis techniques on calcium imaged recordings of neurons, we were well positioned to begin the last part of this thesis, which is the exploratory analysis of glial and neuronal population interaction
- Biomedical Engineering
- Doctor of Philosophy (PhD)