Understanding the role of genetic divergence and recombination in adaptation is crucial to understanding the evolutionary potential of species since they can directly affect the levels of genetic variation present within populations or species. Genetic variation in the functional parts of the genome such as exons or regulatory regions is the raw material for evolution, because natural selection can only operate on phenotypic variation already present in the population. When natural selection acts on a phenotype, it usually results in reduction in the levels of genetic variation at the causal loci, and the surrounding linked loci, due to recombination dynamics (i.e. linkage); the degree to which natural selection influences the genetic differentiation in the linked regions depends on the local recombination rates.

Studies investigating the role of genetic divergence and recombination are common in model species such as Drosophila melanogaster. Only recently have genomic tools allowed us to start investigating their role in shaping genetic variation in non-model species. This thesis adds to the growing research in that domain. In this thesis, I have asked a diverse set of questions to understand the role of genetic divergence and recombination in adaptation in non-model species, with a focus on Lepidoptera.

First, how do we identify causal genetic variation causing adaptive phenotypes? This question is fundamental to evolutionary biology and addressing it requires a well-assembled genome, the generation of which is a cost, labor, and time intensive task. In paper I, I present a tool, MESPA, that stitches together exonic sequences in fragmented assemblies to produce high-quality gene models. These high-quality gene models can be used by researchers in the downstream analyses, providing genomic insights for a fraction of cost of a high quality genome. 

Second, what does the pattern of recombination rate look like in chromosomes that lack centromeres (i.e.holocentric chromosomes)? In paper II, I compare the recombination landscape and the patterns of nucleotide diversity in three Lepidotera with holocentric chromosomes, Pieris napi, Bombyx mandarina, and Bombyx mori, with a monocentric species. Our results show that on average these three Lepidoptera have high rates of recombination across the vast majority of their genome. Our results also suggest that given similar effective population sizes, these species are likely to harbor more genetic diversity compared to monocentric species, which has important evolutionary consequences for these species.

Third, what is the potential for parallelism at the genetic level in convergent melanic phenotypes? In paper III, I investigated the genetic basis of the female-limited melanic phenotype in the green-veined white (Pieris napi) butterfly, and found a 20kb region, approximately 50kb from the gene cortex, associated with this trait. This gene has been implicated in melanic phenotypes in other Lepidoptera that diverged from Pieris approximately 100my, indicating very high predictability for this trait.

Finally, what is the role of cis-regulatory variation in local adaptation? In paper IV, I analyzed the relationship between allele specific expression (ASE) and genetic divergence (F_ST) in the F1 hybrids of Pieris napi napi and Pieris napi adalwinda. I show that intersecting results from ASE with F_ST is a powerful approach to identify genes involved in local adaptation.