Evolutionary and deep mining models for effective biomarker discovery

Alzubaidi, A.H.A., 2019. Evolutionary and deep mining models for effective biomarker discovery. PhD, Nottingham Trent University.

[img]
Preview
Text
PhD_Thesis_AbeerAlzubaidi.pdf

Download (4MB) | Preview

Abstract

With the advent of high-throughput biology, large amounts of molecular data are available for purposeful analysis and evaluation. Extracting relevant knowledge from high-throughput biomedical datasets has become a common goal of current approaches to personalised cancer medicine and understanding cancer genotype and phenotype. However, the datasets are characterised by high dimensionality and relatively small sample sizes with small signal-to-noise ratios. Extracting and interpreting relevant knowledge from such complex datasets therefore remains a significant challenge for the fields of machine learning and data mining. This is evidenced by the limited success these methods have had in detecting robust and reliable biomarkers for cancers and other complicated diseases. This could also explain the lack of finding generic biomarkers among the identified published genes for identical diseases or clinical conditions.

This thesis proposes and evaluates the efficacy of two novel feature mining models established on the basis of the evolutionary computation and deep learning paradigms to position and solve biomarker discovery as an optimisation problem. Deep learning methods lack the transparency and interpretability found in the evolutionary paradigm. To overcome the inherent issue of poor explanatory power associated with the deep learning, this research also introduces a novel deep mining model that helps to deconstruct the internal state of such deep learning models to reveal key determinants underlying its latent representations to aid feature selection. As a result, salient biomarkers for breast cancer and the positivity of the Estrogen and Progesterone receptors are discovered robustly and validated reliably across a wide range of independently generated breast cancer data samples.

Item Type: Thesis
Creators: Alzubaidi, A.H.A.
Date: September 2019
Rights: This work is the intellectual property of the author. You may copy up to 5% of this work for private study, or personal, non-commercial research. Any re-use of the information contained within this document should be fully referenced, quoting the author, title, university, degree level and pagination. Queries or requests for any other use should be directed in the owner of the Intellectual Property Rights.
Divisions: Schools > School of Science and Technology
Depositing User: Linda Sullivan
Date Added: 24 Jun 2020 08:55
Last Modified: 24 Jun 2020 08:55
URI: http://irep.ntu.ac.uk/id/eprint/40093

Actions (login required)

Edit View Edit View

Views

Views per month over past year

Downloads

Downloads per month over past year