UC Santa Cruz

Functional imaging of neural cell populations at mesoscale is criticalfor mapping intra- and inter-regional network dynamics across the dorsal neocortex. Past lab members produced a flexible work flow for producing high quality segmentations of functional structure across neocortex utilizing Independent Component Analysis (ICA). This unsupervised machine learning decomposition of densely sampled recordings of cortical calcium dynamics, results in a collection of components comprised of neuronal signal sources distinct from optical, movement, and vascular artifacts. In this body of work, I built a supervised learning classifier that automatically separates neural activity and artifact components, using a set of extracted spatial and temporal metrics that characterize the respective components. Control data recorded in glial cell reporter and non-fluorescent mouse lines validates human and machine identification of functional component class.

Utilizing the insight from the machine learning analysis, I processed alarge dataset from three different transgenic mouse lines expressing calcium indicators in distinct sub-population of neurons: pan-neuronal, upper, and lower layer specific excitatory neuronal subpopulations. I was able to create domain maps to discretize the cortex into functional units and observe how each domain behaves. The application of this data-driven method ultimately reduced the number of time series, while maintaining the informative structure at a mesoscale. From all these animals, I gain insight into the developing structural changes of circuitry based on age-related functional changes observed in these domain maps.