dc.contributor.author |
Houry, Carine Oussama, |
dc.date.accessioned |
2017-08-30T14:27:33Z |
dc.date.available |
2017-08-30T14:27:33Z |
dc.date.issued |
2016 |
dc.date.submitted |
2016 |
dc.identifier.other |
b19016566 |
dc.identifier.uri |
http://hdl.handle.net/10938/11053 |
dc.description |
Thesis. M.S. American University of Beirut. Department of Biology, 2016. T:6499 |
dc.description |
Advisor : Dr. Heinrich Burggraf Zu Dohna-Schlobi, Assistant Professor, Biology ; Committee members : Dr. Khouzama Knio, Professor, Biology ; Dr. Zakaria Kambris, Assistant Professor, Biology. |
dc.description |
Includes bibliographical references (leaves 29-34) |
dc.description.abstract |
The endosymbiotic bacterium Wolbachia infects wide range of arthropods and their relatives. It is an intracellular parasite transmitted through the egg from mother to offspring. Wolbachia can spread and persist through various means of host reproductive manipulation. These manipulations include cytoplasmic incompatibility, the selective killing of male offspring, induction of parthenogenesis, feminization of genetic males. It is unclear to what degree the induced host phenotype is determined by Wolbachia genetics and how Wolbachia evolved such a wide array of host manipulations. We assessed the phylogenetic signal within our studied trait, i.e. the Wolbachia-induced host phenotype, by comparing the likelihood of a model based on a star tree to the likelihood of the model based on our actual tree. Our results suggest that the model based on the actual tree has the higher likelihood inferring that the Wolbachia-induced host phenotype is dependent on the Wolbachia strain. To understand transition rates from one host phenotype to another, we fitted contrasting models for evolutionary transitions to a phylogenetic tree and determined which one is more consistent with data. We used two model selection procedures: Bayes factor approximation from harmonic mean and model likelihood comparisons with optimizing parameters. The former approach favors a model with no constraints imposed on the data oppositely to the latter approach which favors a model with a single transition rate from and to all host phenotypes. Our results suggest that the estimations based on maximum likelihood are more conclusive, in view of drawbacks of Bayes factor approximation from harmonic mean and maximum parsimony considerations. |
dc.format.extent |
1 online resource (x, 34 leaves) : illustrations (some color) |
dc.language.iso |
eng |
dc.relation.ispartof |
Theses, Dissertations, and Projects |
dc.subject.classification |
T:006499 |
dc.subject.lcsh |
Insects -- Pathogens. |
dc.subject.lcsh |
Wolbachia. |
dc.subject.lcsh |
Phylogeny. |
dc.subject.lcsh |
Arthropoda -- Microbiology. |
dc.subject.lcsh |
Bioinformatics. |
dc.subject.lcsh |
Computational biology. |
dc.subject.lcsh |
Genomics. |
dc.title |
Investigating evolutionary transitions in Wolbachia pipientis’ propensity to induce different reproductive host phenotypes - |
dc.type |
Thesis |
dc.contributor.department |
Faculty of Arts and Sciences. |
dc.contributor.department |
Department of Biology, |
dc.contributor.institution |
American University of Beirut. |