Dynamic gene regulatory network construction from high-throughput time-course data

The rapid advancement of high-throughput genomic technologies has created many opportunities to analyze gene expression and gain insights into complex biological processes. In particular, time-course gene expression data has become important for understanding the dynamic response of biological systems and for constructing and analyzing dynamic gene regulatory networks (GRNs) that denote the interaction between genes. However, the analysis of time-course gene expression data presents numerous statistical and computational challenges due to the high dimensionality of the data and substantial measurement error. This thesis addresses several of these challenges in the context of constructing and analyzing GRNs by developing four novel statistical methodologies aimed at improving the pre-processing of time-course data, clustering in both temporal and spatial contexts, and the statistical analysis of samples of GRNs.