Date of Award

2010

Document Type

Master Thesis

Degree Name

Masters of Science (Research)

Department

Computing

First Advisor

Dr. Pio Fenton

Second Advisor

David Simpson

Abstract

The manual interpretation of mass spectra is a complex and time consuming task. The problem of manually interpreting this data is further exacerbated by the large numbers of mass spectra which can potentially be produced in a single proteoniics experiment. This shows the need for high throughput approaches to the interpretation of mass spectra. Existing automated approaches are however error prone due to the complexity of the task. Accordingly, this thesis discusses and evaluates the application of neural networks to improving the sensitivity, specificity and robustness of current approaches to the automated interpretation of such mass spectral data.

Several neural network approaches and architectures are evaluated in an attempt to define a neural network suited to the interpretation of the mass spectral data of peptides. The best performing networks are evaluated on their potential to offer a lift in performance to elements of the current non-neural network based approaches to the interpretation of the mass spectra of peptides. A novel automated peptide sequencing application is developed which relies heavily on the output produced by the neural network at all stages of its process. This application allows for an assessment of the capability of neural networks in the domain and the identification of tasks to which neural networks offer superior performance.

The neural networks developed in this research attain high levels of performance and robustness. Good results were achieved in terms of both sensitivity and specificity on the task of mass spectral peak classification. Despite the overall good performance attained for the neural network, generally inferior performance was observed for the novel peptide sequencing application when benchmarked against classical (non-neural network based) automated approaches. The results attained are however promising with several instances observed of superior performance to the PepNovo sequencing application.

Access Level

info:eu-repo/semantics/openAccess

Download

Share

COinS