One of the driving goals of the field of comparative genomics is to understand the amount of variation within and between genomes, and how this variation can produce the broad array of phenotypes seen across the tree of life. Recent developments in sequencing and assembly have enabled researchers to explore most any species of interest in detail. In my dissertation work, I exploit these developments to generate high quality genome assemblies for the species of the Drosophila yakuba clade of vinegar flies, and explore variation within and between the species. As a prerequisite to any investigation into genetic variation in non-model species, I examine sources of genotyping error through simulations and identify filtering criteria to mitigate them. Having multiple high quality assemblies of a single species enables validation of these filtering criteria using real data. I then use these genome assemblies in concert with population resequencing datasets for each species to understand patterns of nucleotide diversity in D. santomea, D. yakuba, and D. teissieri, especially as it pertains to constraint, efficacy of purifying selection, and the distribution of fitness effects of new mutations. Finally, I explore the realm of structural variation within the D. yakuba genome, characterizing four segregating inversions of young to intermediate age. These inversions have strong structuring effects on nucleotide diversity, and lead to widespread changes in gene expression. In this process of studying inversions from the ground up, I identify genes that are both differentially and allele-specifically expressed between inversion arrangements, and find that two of these genes are related to response to thermal stimuli, suggesting temperature preference as a potential phenotypic consequence of these inversions. Through these chapters, my dissertation work shows the power of modern comparative genomics to understand variation in non-model species.