Development of a statistical tool for comparative analysis of gene expression dynamics and its application using Brassica and Arabidopsis transcriptomic data

Kristianingsih, Ruth (2025) Development of a statistical tool for comparative analysis of gene expression dynamics and its application using Brassica and Arabidopsis transcriptomic data. Doctoral thesis, University of East Anglia.

Preview

PDF Download (12MB) | Preview

Abstract

Comparing gene expression patterns can help identify correspondences of developmental stages within and between species, highlight differences in the timing of key developmental events, and elucidate transcriptional responses to treatments. However, such comparisons are often complicated by variations in timing and the differing timescales of these events. To overcome this challenge, we developed a method based on curve registration, which optimally aligns gene expression dynamics by inferring temporal shifts and stretches. Statistical evaluation of these parameters allows us to compare the fit of a non-registered model (in which expression profiles are considered different) against a registered model (in which differences are resolved through alignment). To make this approach widely accessible, we implemented it as an R package, greatR. This tool has been validated using various datasets, both simulated datasets and real biological data. greatR has been successfully applied to multiple comparisons, including the floral transition in Arabidopsis, B. rapa, and B. oleracea, as well as across these species. Additionally, we employed greatR to compare expression profiles of two Arabidopsis genotypes during bract formation, offering new insights into the genetic and transcriptional mechanisms underlying this trait. Beyond plant systems, greatR can be extended to compare expression responses in other organisms, making it a valuable tool for cross-species analysis. greatR has proven to be able to detect pairs of genes with expression profiles which can be superimposed and, therefore, have similar dynamics. This approach enables the exploration of dynamic differences in gene expression within and across species, providing an important foundation for understanding the regulatory networks that govern various biological processes. By comparing these dynamics, it can help uncover both conserved and species-specific regulatory mechanisms. This approach facilitates the transfer of knowledge from well-studied model organisms to less-explored species, the identification of co-regulated gene modules, and the discovery of temporally differentially expressed genes linked to specific conditions or traits.

Item Type:	Thesis (Doctoral)
Faculty \ School:	Faculty of Science > School of Biological Sciences
Depositing User:	Chris White
Date Deposited:	03 Jul 2025 09:28
Last Modified:	03 Jul 2025 09:28
URI:	https://ueaeprints.uea.ac.uk/id/eprint/99822
DOI:

Downloads

Actions (login required)

View Item