Research

Spatial Genetics of Beach Moles (Haustoriidae: Amphipoda)

​

Most organisms are dispersal-limited, and this generates patterns of isolation-by-distance with respect to genetic ancestry. The degree of population structure is shaped by the ratio of an organism's dispersal ability to the size of its range. When this ratio is very small, population structure is strong, and genetic ancestry becomes correlated with distance.

​

Beach moles (Haustoriidae: Amphipoda) live on sandy beaches across the Northern Hemisphere. My research focuses on one species, the Canadian beach mole (Haustorius canadensis), which ranges from southern Florida to southern Maine. Despite this incredible range that covers a dozen degrees of latitude, beach moles are blind, burrowing organisms that rarely swim, and keep their offspring in little brood pouches (called marsupium). This severely limits their dispersal ability, and this pattern is reflected in the strong correlation between individual genetic ancestry and geographic space.

​

My research on beach moles aims to examine long-standing hypotheses from population genetics and landscape ecology. In both, there exists an interest in how range edges shape ancestry with a general idea that range peripheries have lower population densities than range centers. Since beach moles live in well-defined, one-dimensional habitats, we can directly examine this hypothesis. Furthermore, my research seeks to understand how wave currents interact with dispersal ability to shape migration trajectories in coastal organisms. Lastly, my research hopes to investigate how mixed generational structure and parental investment interact with genetic load in large populations. 

​

SANDWRM: Spatial Population Genetic Methods

​

Given the ubiquity of spatial structure in natural systems, there is a pressing need for new statistical methods that can account for the correlation between geography and ancestry. In the past, methods have been developed that require arbitrarily discretizing the individuals into randomly mating units and then calculating the degree of genetic differentiation between these. Increasingly, researchers have sample locations across broad areas, which have uncovered continuous patterns of differentiation. Thus, ideally there would exist statistical methods that treat individuals as the unit of measurement instead of subpopulations (as with the F-statistics). Furthermore, there is a pressing need, especially in human genetics, to abandon arbitrary geographic designations. 

​

Inspired by these needs, during my postdoc I developed a method for estimating Wright's neighborhood size by fitting isolation-by-distance curves. In this way, researchers need only supply pairwise geographic and genetic distance matrices, without the need to designate individuals into subpopulations. Furthermore, the method jointly infers the long-term effective population size, permitting investigators to partition the population pedigree into recent and deep-time events.

​

The method is current being developed into an R package called SANDWRM (Spatial Analysis of Neighborhood size and Diversity using the WRight-Malecot model). In the future, I'd like to expand this model into a phylogenetic context, permitting researchers to incorporate continuous space into phylogenetic inference as well. 

​

Phylogenomics of Amphipoda

​

Amphipod crustaceans are one of the most diverse groups of arthropods in the ocean with ~25,000 species. Despite their great diversity, we know little about their relationships, with many taxonomic guides listing families in alphabetical order instead of by their phylogenetic affinity. 

​

My research seeks to incorporate cutting-edge statistical and genomic sequencing technologies to generate an exhaustive phylogeny of Amphipoda. 

​

Evolution of Giant Genomes

​

Genome size varies by orders of magnitude across the Tree of Life and lacks any correlation with organismal complexity (called the C-value paradox). Beach moles, despite being only 4–15mm in length, have gigantic genomes; the Canadian beach mole, for example, has a genome ~13 Gb, nearly 4x the size of a humans. Why?

​

Research into the evolution of large genomes tends to focus on the emergence of repetitive and parasitic DNA. In some beach mole species, they harbor >70% parasitic DNA. What are the evolutionary forces that permit this rapid accumulation of transposons? Why does natural selection allow this to occur? My research seeks to answer these questions using both simulation-based and empirical reconstructions of transposable element evolution, focusing on the mutation-hazard hypothesis to explain the proliferation of parasitic DNA.