Files
Abstract
Although many programs exist for generating simulated alignments of DNA sequences, no pro-gram combines the robust GTR substitution model with a model of indel formation. To correct this gap, the application, Dawg, was created. Dawg is the first program to provide the option of producing indels under the power-law model, which is the model most consistent with biological data.Some authors have suggested that the power-law distribution of indel lengths make logarithmic gap costs the preferred method of sequence alignment. Utilizing Dawg, the utility of logarithmic gap costs is investigated in pairwise global alignments. They are shown to be poor performers when compared to the standard affine methods. Furthermore, a model for calculating gap costs is developed which explains why affine costs are a better option than logarithmic costs.The third study investigates the antagonism between local dispersal and self-incompatibility systems in a continuous plant population. This antagonism is shown to affect the fine-scaled genetic structure of the population and depend on the linkage of markers and self-incompatibility loci.Furthermore, gametophytic and sporophytic self-incompatibility are show to not be much different than obligate outcrossing.