Project abstract for group reillyc

High-Dimensional Nonparametric Longitudinal Analysis

The goal of this project is to develop flexible methods for the analysis of high dimensional data (e.g. RNA-Seq or metabolomic data) that is collected according to a longitudinal sample collection protocol. The researchers use a level crossing approach that combines information across the levels using Bayesian non-parametric models that use a mixture of Dirichlet process priors. They use Markov chain Monte Carlo methods to conduct inference.