Peter Gui

Date of Award

2017

Document Type

Thesis

Degree Name

Bachelors

Department

Natural Sciences

First Advisor

Ryba, Tyrone

Area of Concentration

Computational Biology

Abstract

Network structures can be used to analyze both nucleotide sequence and epigenetic information. By creating network structures which can be programmed to analyze information and change their shape over a given sequence we can create a framework for mining and consolidating bioinformatic data about a given locus within the genome of a species. This framework will allow bioinformaticians to easily apply statistical models and machine learning algorithms to the genome at various levels in order to hierarchically abstract the information that we are receiving from the genome with its four basic nucleotides. My goal for this thesis was to begin experimenting with networks and sequence data to figure out what types of algorithms are useful for analyzing these data types. Networks, specifically, are used because they allow a programmer to not only look at the amino acid sequence associated with the locus, but also view the components that make up that data structure. Furthermore this data structure can be combined with other networks to build higher level objects. My hope is that this theory could accelerate the process of data mining by standardizing an object type which holds any information one could posit about a DNA sequence. With this relatively simple and standard data type other users will be able to build even higher level objects from existing objects in a hierarchical manner. This thesis will cover the progress that I made both in understanding the theory of networks and in the implementation of algorithms for managing these object types. As I began to experiment with network objects I came to realize that they could be useful for a large number of seemingly disperate applications within bioinformatics. Therefor the primary goal of this thesis is to support the argument that networks and network based algorithms could help make parts bioinformatics more coherent, user friendly and autonomous.

Download

Share

COinS