Reference Graph Construction and Merging for Human Genomic Sequences
Year: 2016
Type: dissertation
Abstract: The focus of this thesis is on the computational challenges faced when reading out human DNA. In particular, the reads of the DNA are commonly aligned with the help of a reference string. In this thesis, the opportunities and difficulties when using a graph reference instead are explored. \nSeveral approaches for using graph references have already been proposed, but none of them are used widely in this field. Therefore here we compare the existing approaches and develop new algorithms for working with genomic graphs. This is intended to further help spread the usage of graph references. In particular, we develop algorithms for merging and fusing flat XBW tables, which represent genomic graphs. We implement these algorithms in the Graph Merging Library GML, which can be used to visualize their behaviour to further the understanding of these methods. In addition to being used for the alignment of reads to reference graphs, genomic graphs can also arise in other situations. Our newly developed algorithms are general enough to also be of use in these scenarios. \nOverall, this thesis constitutes a step in the direction of adopting the more complicated but more powerful graphs rather than sequential string data within the field of DNA sequencing.
Source: Skemman
Author: Tom Willy Schiller
Institution: University of Iceland
Language: English
Subfield: Theoretical Computer Science
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2289922653
