Rebecka Jörnsten, Department of Statistics, Rutgers University

 
Statistical analysis of dendritic branching in hippocampal neurons

Abstract: The analysis of neuron morphology is geared toward
increasing our understanding of dendritic branching and developing
mechanisms.  Neurological disorders such as autism and Rett's syndrome
are associated with disrupted or altered branching
patterns. Understanding which factors affect branching and how they do
so is the first step toward developing drugs that can target these
disruptions directly.  We analyze branching patterns of neurons from a
series of experiments via  multivariate mixtures of generalized linear
models. The data  consists of primary dendrites (stemming from the
cell body) and their first level of branching into secondary
dendrites. We jointly analyze which factors affect the number of
primary dendrites, the proportions of primaries that branch, and how
many secondaries result from a branching event. The outcome of these
analyses enabled us to formulate novel biological hypotheses that are
currently being validated in the lab.