Dissertation
Computational methods for estimating genetic relationships
Doctor of Philosophy (PhD), Washington State University
01/2012
Handle:
https://hdl.handle.net/2376/4264
Abstract
The advent of molecular methods has altered our approach to the study of the genetic relationships of microbes. In particular, we now have the unprecedented ability to estimate genetic relationships from whole genome sequences whose numbers are increasing exponentially. In this dissertation we examine several computational methods for using genome sequences to infer the genetic relationships of both plasmids and bacteria. First we describe a method for relating 527 Gram-negative bacterial plasmids based on their genetic sequences. Initial classification of their genetic relationships was accomplished using a computational approach analogous to hybridization of "mixed-genome microarrays." Relationships were refined for several clusters by identifying conserved proteins within a cluster. The replication of consistent results produced in a separate study for a small group of IncA/C plasmids and clusters of Borrelia plasmids provides evidence that the approach used can correctly predict genetic relationships. Second, we use the pClust program to estimate the genetic relationships of the same 527 plasmid genomes. Protein clusters generated by pClust are used to create profiles for each plasmid in the tree, which are then used as correlation filters for classification of a new bacterial plasmid. The major contribution of this work is the development of a method that can be used to construct a tree and, more importantly, to insert a new taxon a posteriori. While this method was developed specifically for plasmids, it can be used with genomes of any kind. The third project is a study of the genetic relationships of bacteria, more specifically species of the alphaproteobacteria class. Typically phylogeny studies of bacteria are based on the 16S rRNA gene. In this work, however, we again use the software program pClust with twelve genomes to generate homologous protein clusters which are then used to construct a tree. The results are compared with a tree constructed using 16S rRNA; while certain features in both trees are similar, the differences indi- cate that the use of whole-genome sequences may provide a better estimate of genetic relationships.
Metrics
5 File views/ downloads
19 Record Views
Details
- Title
- Computational methods for estimating genetic relationships
- Creators
- Yunyun Zhou
- Contributors
- Shira L. Broschat (Advisor)Douglas R. Call (Committee Member)John H. Miller (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 109
- Identifiers
- 99900581543601842
- Language
- English
- Resource Type
- Dissertation