Return to search

Estimating and modeling gene flow for a spatially distributed species

We model the genetic behavior of a large population of individuals which is to some extent subdivided into colonies. At some stage of the life cycle there is genetic exchange among the colonies due to migration. We consider the stationary distribution of allele frequencies at a specific locus over all the colonies under various models of migration among the colonies. All of our models are variations of the Wright-Fisher Model. We consider both estimating and modeling gene flow among the colonies. / The Island Model assumes that migrants into a given colony are equally likely to come from any of the other colonies. In the Island Model with all mutations producing new alleles (the infinite-allele model), we use a moment argument to provide analytic support that the Beta (or Dirichlet) distribution is a good model for the stationary distribution of allele frequencies. In simulated data, we compare the performance of several candidate estimators of migration rate. / If migration is more likely to be from nearest-neighbor colonies, then the Stepping-Stone Model is an appropriate extension of the Island Model. For a stepping-stone migration pattern we consider the two-allele case, and study the relative frequency of allele A1 over all the colonies. The effects of various migration patterns across colonies on these relative frequencies are studied. We set down a joint distribution for the relative frequencies of A1 at the colonies. This joint distribution is Gaussian and allows us to estimate parameters that describe the extent of genetic exchange across colonies. The Gaussian Model is easy to simulate and fits well to data (possibly transformed) that has been generated according to the Stepping-Stone Model. The Gaussian Model can be extended without much difficulty to describe various mating patterns across colonies. / Source: Dissertation Abstracts International, Volume: 53-07, Section: B, page: 3252. / Major Professor: Duane Meeter. / Thesis (Ph.D.)--The Florida State University, 1992.

Identiferoai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_76700
ContributorsBurr, Thomas Lee., Florida State University
Source SetsFlorida State University
LanguageEnglish
Detected LanguageEnglish
TypeText
Format101 p.
RightsOn campus use only.
RelationDissertation Abstracts International

Page generated in 0.0017 seconds