Inferences from nucleotide sequences: Recombination, haplotypes and rates of molecular evolution

The advent of nucleotide sequencing has provided a wealth of information for biologists. This dissertation illustrates how population genetics and molecular evolution theory can inform the extraction of rich historical information latent in nucleotide sequences.; Resequenced regions of DNA in a common sample of individuals from wild barley were used to estimate the relative contribution of meiotic recombination and mutation on the genetic diversity of the population. Mutation introduces new variants in a population, ultimately rearranged by the process of recombination. Recombination occurs through one of two processes during meiosis. One process is a reciprocal exchange at one point on a chromosome, called crossing over. The other is homologous gene conversion in which a small tract of DNA is exchanged. Accurately estimating the extent of recombination and specifically estimating homologous gene conversion depends on very accurate data. In diverse, outcrossing diploid organisms, sequence accuracy often depends on the separation of the sequences on the two homologous chromosomes, called the haplotypes.; A method is described for the detection and correction of genotyping errors in haplotype data by considering the recombinational history implied by the patterns of variation, three SNPs at a time. The method is implemented in a perl script called EDUT (Error Detection Using Triplets).; Homologous regions of nucleotide sequence collected in a sample of multiple related species provide information about a larger scale of evolutionary time. The rates of evolution for a set of ten homologous regions in a common set of morning glory species were estimated, taking advantage of the shared history of the nucleotide regions. Partitions of the data were considered for hypothesis testing. The partitions include each of the nineteen species, groups of related species, the ten genes sequenced, and three categories of genes sequenced. The ten genes include genes coding for enzymes taking part in two different biochemical pathways and for one regulatory MADS box gene. The effects of each gene and lineage on synonymous, nonsynonymous and overall rates of nucleotide change were considered, leading to a conclusion that mutation rate variation may explain the patterns of rate variation.