Identification of prognostic indicators of healthy and unhealthy conditions with a machine learning-based systems biology approach using gut microbiome data

Madgwick, Matthew (2022) Identification of prognostic indicators of healthy and unhealthy conditions with a machine learning-based systems biology approach using gut microbiome data. Doctoral thesis, University of East Anglia.

Preview

PDF Download (35MB) | Preview

Abstract

Inflammatory bowel disease (IBD) is associated with alterations in the intestinal microbiome. However, the precise nature of these microbial changes remains unclear. With billions of microbes within the gut, novel and powerful computational techniques are required to identify the relevant shifts in the microbiota that contribute to healthy and unhealthy conditions.

Machine learning (ML) allows a data-driven approach to identify these discrete dynamic changes. However, the interpretation and biological validation of the findings from ML algorithms remain a challenge. By combining ML and Systems Biology (SB) approaches, this thesis aims to characterise key microbial factors in IBD pathogenesis by extracting prognostic indicators from the human gut microbiome.

The causal relationship between the changes in the gut microbiome and IBD is difficult to establish. Data from cross-sectional studies are plagued by confounding factors and inconsistencies between cohorts. Rich longitudinal datasets and integrated metagenomic, multi-omic, and electronic healthcare records can be used to overcome these limitations. In this PhD thesis, I have developed an integrated ML-based microbiome analysis pipeline to identify prognostic indicators for IBD from longitudinal microbiome data. Furthermore, using a variety of SB approaches, the interplay between the host and the microbiome has been explored to provide insights into the mechanisms during healthy and unhealthy conditions.

Item Type:	Thesis (Doctoral)
Faculty \ School:	Faculty of Science > School of Biological Sciences
Depositing User:	Chris White
Date Deposited:	28 Feb 2024 10:38
Last Modified:	28 Feb 2024 10:38
URI:	https://ueaeprints.uea.ac.uk/id/eprint/94370
DOI:

Downloads

Actions (login required)

View Item