Full text
| |
| Author(s): |
Flavia Fuchs de Jesus
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Campinas, SP. |
| Institution: | Universidade Estadual de Campinas (UNICAMP). Instituto de Biologia |
| Defense date: | 2006-08-25 |
| Examining board members: |
Vera Nisaka Solferini;
Louis Bernard Klaczko;
Sérgio Furtado dos Reis;
Paulo Alberto Otto;
Enrique Pablo Lessa Gallinal;
Aluísio de Souza Pinheiro
|
| Advisor: | John Wakeley; Vera Nisaka Solferini |
| Abstract | |
Population structure is one of the major factors shaping the pattems of genetic variation across time and space. Due to the climatic fluctuations of the Qua terna ry, several species may have suffered population reduction and fragmentation, becoming restricted to refugia during glacial periods and expanding again during interglacials. This has been used to explain some patterns currently observed in several species. The present work consisted in the development and study of models to help understand the genetic consequences of cyclic changes in population structure and size, such as the ones that may have occurred throughout the climatic fluctuations of the Quatemary. Population reduction may cause reduction in population effective size, mean coalescence time and genetic variation; whereas an increase in population subdivision may have the opposite effect. In order to investigate these two opposite effects, two models were studied, both with two alternating phases, corresponding to the glacial and interglacial periods. Both models included changes in population structure, besides those in population size, in a cyclic manner. In the first model, completely panrnictic phases were alternated with completely structured ones. Based on this model, an expression was derived for the expectation of coalescence times of two sequences and, from this, an expression for the expectation of the number of segregating sites. Both an increase in the number of demes and in the duration of the structured phases caused an increase in coalescence times and levels of genetic variation. The results obtained were compared to what would be expected for a panrnictic population of constant size. It was verified that population structure may outhweigh the effect of population reduction during glacial periods. Specifically, the mean number of segregating sites can be greater in the proposed model, even when population size is quite reduced during the structured phases. In the second mode!, population subdivision was allowed in both phases' - according the finite island model with migration. Population size, migration rate and number of demes varied between phases. For this model, besides an expression for the mean coalescence time, an expression for the distribution of coalescence times was also obtained. The distributions observed were quite different from what would be expected for a panrnictic population of constant size. Population reduction during glacial periods caused discontinuities and multiple peaks in the distribution of coalescence times, as well as a reduction in the expected times. An increase in population structure, through reducing migration rates, increased the mean times and attenuated the peaks of the distribution. Mean coalescence times, in general, also increased with a greater number of demes during glacial periods. The results obtained help understand the genetic consequences of glacial cycles, and, especially, point to the importance of population structure for the maintenance of genetic varlation. Besides, they offer a potential explanation for the genetic pattems observed in several species, for which long gene genealogies are observed, with the most recent ancestor predating by far the last glacial period (AU) | |