|Institution:||University of Washington|
|Keywords:||Bipartite; Graph theory; Inference; Networks; Neuroscience; Rhythms; Mathematics; Neurosciences; Statistics; Applied mathematics|
|Full text PDF:||http://hdl.handle.net/1773/40831|
At first glance, the neuronal network seems like a tangled web in many areas throughout the nervous system. Often, our best guess is that such messy connections are close to random, while obeying certain statistical constraints, e.g. the number of connections per neuron. However, neuronal wiring is coordinated across larger mesoscopic distances in a way that differentiates between brain layers, areas, and groups of cells. We work across spatial scales in order to understand this hierarchy of order and disorder in brain networks. Ultimately, the goal is to understand how network structure is important for brain function. This leads to: 1. An inference technique which reconstructs mesoscopic brain networks from tracing experiments targeting spatially contiguous groups of neurons. 2. Models of networks which are random, while also having constrained average connectivity and group structure. 3. Comparing simulated and real respiratory rhythms, highlighting the role of inhibitory neurons and connectivity on rhythmogenesis, in particular synchrony and irregularity.Advisors/Committee Members: Shea-Brown, Eric (advisor).