Use this URL to cite or link to this record in EThOS:
Title: Machine learning based approaches for identifying sarcopenia-related genomic biomarkers in ageing males
Author: Dreder, Abdouladeem
ISNI:       0000 0004 7430 0743
Awarding Body: Northumbria University
Current Institution: Northumbria University
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Sexual dimorphism of skeletal muscle can occur due to age and many of these age-related changes in skeletal muscle appear to be influenced by gender. In humans, the muscle mass peaks in the second decade while loss of muscle mass (sarcopenia) starts between the third and the fifth decade of life. In system biology, the function of genes still needs to be understood and understanding gene function remains a significant challenge. Several machine learning and computational techniques have been used to understand. However, these previous attempts have not produced enough interpretation of the impact of age on skeletal muscle mass across both gender. Although there are several thousands of genes, very few differentially expressed genes play an active role in understanding the age and gender differences. The core aim of this thesis is to uncover new biomarkers that can contribute towards the prevention of sarcopenia progress in humans according to the gene expression levels of skeletal muscle tissues. The main contributions are the development of machine learning methods based on majority voting of multi-evaluation methods and multi-feature selection methods in order to analyse microarray data and identify subsets of genes related to muscle mass loss in ageing males and females. Previously, statistical methods were used to find important genes related to the impact of age on muscle mass loss. Multi-filter and multi- wrapper based systems are proposed in this thesis to identify different and common sarcopenia-related genes in males and females based on human skeletal muscle. Genes are first sorted using three different evaluation methods (t-test, Entropy and Receiver operating characteristic). Then, important genes are obtained using majority voting based on the principle that combining multiple models can improve the generalization of the system. Experiments were conducted on three different microarray gene expression datasets and results have indicated a significant increase in classification accuracy up to 10% associated with sarcopenia when compared with existing systems.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: B900 Others in Subjects allied to Medicine ; G900 Others in Mathematical and Computing Sciences