Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Arts and Sciences



First Advisor

Dr. Catherine R. Linnen


To fully understand the genetic basis of adaptation, we need to know its predictability—the extent to which specific selective pressures and contexts can yield corresponding genetic changes. In particular, the repeated colonization of similar, specialized environments by different taxa is ideal for assessing the frequency of reoccurring changes in the same genes or functions. But compared to a growing body of literature on the convergent evolution of individual genes, far less is known about the repeatability of gene family evolution, where families (defined here as groups of genes that share sequence and functional similarity from common ancestry) can expand (gain genes) or contract (lose genes) in response to specific selective pressures.

Here, I compared candidate gene families among closely related taxa that vary in ecology, namely diet breadth in the lecontei subgenus of Neodiprion sawflies (Hymenoptera: Diprionidae). This is a monophyletic group of insects that shifted from angiosperm plants to pine trees within the last ~60 mya. Underlying these comparisons is a large body of work in Drosophila that identified genetic changes in chemoreceptor genes associated with a shift to an increasingly specialized lifestyle: neutral or adaptive decay (loss-of-function mutations and missing genes), gene family expansions, and positive selection on individual genes.

To determine the generality of these signatures and family expansion/contraction, I described these patterns in other gene families and at multiple levels of biological organization within the lecontei subgenus. First, I compared chemosensory, detoxification, and immunity families within the same genome; these are subject to similar selective pressures and population history. I also compared these families across the order Hymenoptera to identify associations between gene family size and eusociality, specialist or generalist status, or diet. Second, I compared the chemoreceptor gene family in a pair of sister taxa with pronounced differences in diet breadth. Relative to N. lecontei, which feeds on multiple pine species, N. lecontei is a specialist that uses a single pine species. Third, I compared multiple generalist and specialist sister taxa pairs in the lecontei subgenus to determine whether the patterns identified in the N. lecontei and N. lecontei comparison hold across the subgenus.

Overall, my dissertation research has improved our understanding of evolutionary predictability, or the lack thereof, in several ways. First, gene families appear to vary in both temporal dynamics (shallow vs. deep divergence times) and ecological drivers of size change. In Hymenoptera, social behavior, not diet, has a predictable impact on chemoreceptor family size. Second, despite pronounced differences in diet, N. lecontei and N. lecontei had surprisingly minimal differences in chemoreceptor gene family size or selective history. This suggests that the signatures found in Drosophila do not extend to other taxa. Third, my research is revealing genes and gene families associated with pine feeding and adaptation to specific pine species. Together, these results from multiple levels of biological organization reveal factors that may impact the predictability of gene family evolution.

Digital Object Identifier (DOI)

Funding Information

USDA (United States Department of Agriculture): 2016-2019

KSEF (Kentucky Science & Engineering Foundation): 2015