Please note: this collection is no longer updated. Visit our Theses and Dissertations Collection in VERSO for all U of I ETD since 2012.
ETD RECORD
Likelihood methods to infer balancing selection under K-allele models
Citation
Buzbas, Erkan Ozge.. (2009). Likelihood methods to infer balancing selection under K-allele models. Theses and Dissertations Collection, University of Idaho Library Digital Collections. https://www.lib.uidaho.edu/digital/etd/items/etd_43.html
- Title:
- Likelihood methods to infer balancing selection under K-allele models
- Author:
- Buzbas, Erkan Ozge.
- Date:
- 2009
- Keywords:
- Mutation (Biology)--Mathematical models
- Program:
- Bioinformatics & Computational Biology
- Abstract:
- A balanced pattern in the allele frequencies of polymorphic loci is a potential sign of selection, particularly of overdominance. Although this type of selection is of some interest in population genetics, there exist no likelihood based approaches specifically tailored to make inference on selection intensity. To fill this gap, we present likelihood methods to estimate selection intensity under k-allele models with overdominance.;The stationary distribution of allele frequencies under a variety of Wright-Fisher k-allele models with selection and parent independent mutation is well studied. However, the statistical properties of maximum likelihood estimates of parameters under these models are not well understood. We show that under each of these models, there is a point in data space which carries the strongest possible signal for selection, yet, at this point, the likelihood is unbounded. This result remains valid even if all of the mutation parameters are assumed to be known. Therefore, standard simulation approaches used to approximate the sampling distribution of the maximum likelihood estimate produce numerically unstable results in the presence of substantial selection.;We describe the Bayesian alternative where the posterior distribution tends to produce more accurate and reliable interval estimates for the selection intensity at a locus. In particular, we present methods for single locus and multiple loci k -allele models, including a case where there is epistasis between loci. For the multiple loci model without epistasis, we assume a hierarchical setup between loci to estimate the posterior distribution of the mean selection intensity in a multi locus region of the genome. For the epistatic case, we focus on two types of epistasis: synergistic (antagonistic), where the fitness of the genotype decreases more (less) severely in comparison to the case of independence between loci. We estimate the posterior distribution of the selection intensity for a group of epistatically interacting loci using recent theoretical developments. We provide methods to generate data and to test for independence between loci under selection. Simulated data are used to validate the methods and real data at the Human Leukocyte Antigen loci are analyzed to illustrate an application.
- Description:
- Thesis (Ph. D., Bioinformatics and Computational Biology)--University of Idaho, May 2009.
- Major Professor:
- Paul Joyce.
- Defense Date:
- May 2009.
- Type:
- Text
- Format Original:
- x, 73 leaves :ill. ;29 cm.
- Format:
- record
Rights
- Rights:
- In Copyright - Educational Use Permitted. For more information, please contact University of Idaho Library Special Collections and Archives Department at libspec@uidaho.edu.
- Standardized Rights:
- http://rightsstatements.org/vocab/InC-EDU/1.0/