Research

R Package for Species-Area Relationships

My R package SSARP (Species-/Speciation-Area Relationship Projector) provides a suite of functions to help users create speciation- and species-area relationships for island-dwelling taxa using occurrence data from . The goal of this R package is to help island biogeographers create species-area relationships more easily and learn more about biodiversity patterns in their systems. At the time of writing this description, SSARP can be used to create species-area relationships, but not yet speciation-area relationships. Learn more about the functions included within the package by

Using SSARP to Explore the Impact of Non-Native Species on Species-Area Relationships

I love thinking about the processes that lead to adaptive radiations, especially those that have occurred in squamates. I am, of course, drawn to the classic example of Anolis lizards, but other genera such as Phelsuma and Emoia have also experienced adaptive radiations. Another key feature of these systems is the accidental sharing of lizards between islands via boats and other human-related movement. Lizards that arrive on new islands often easily adapt to their new homes due to similar environments across archipelagoes, and sometimes non-native lizards displace the native species. When creating species-area relationships to visualize biodiversity patterns across archipelagoes for a given genus, how often do researchers filter their data to ensure that non-native species are not skewing the data? If we compare a species-area relationship that only includes native occurrences to one that includes both native and non-native occurrences, does the relationship change? To answer these questions, I am using my R package SSARP (described above) to gather and filter occurrence data for squamate genera that live on islands around the world. I have found that the relationships absolutely change when non-native species are added. For example, the slopes in the species-area relationship for Anolis get flatter and the breakpoint moves forward when non-native species are added.

Arbor Workflows

The goal of web applications built by the Arbor Workflows team is to create more accessible tools for studying phylogenetics. These applications simply require the user to choose a method of interest, upload their data, and press “Go” to complete phylogenetic analyses. My role on the Arbor Workflows team is to create new, intuitive web-based tools to allow users to easily perform phylogenetic comparative methods. Please feel free to .

Additional notes for those interested in how Arbor is built: data management is handled by and its REST API allows us to call methods within Python scripts and receive results from the analyses conducted within those scripts. In many cases, I write the code necessary for a particular phylogenetic comparative method in R, which is then wrapped in Python using . Once a method is called by the user interface, it runs in the background and the results are served back through the API to the user interface page (often as a data table or an associated plot).

Exploring Evolutionary Predictability Using Video Games

Project Hastur is an evolutionary tower defense video game developed by the at the University of Idaho. In this game, the enemies evolve to combat the player’s strategy through a mathematical model of biological evolution. The fitness of enemies is based on how long they survive in the fight against the player, and enemies with the highest fitness reproduce at the end of each generation. This evolution of enemies makes the game more difficult for players if they aren’t paying attention. For example, if the player uses mostly turrets that shoot fire bullets, the enemies will eventually evolve to be resistant to fire bullets. This player will need to change their strategy and use a different strategy if they hope to beat future generations of enemies.

Taking this evolution into account, Project Hastur provides the perfect playground for studies about evolutionary concepts. When different players use the same strategy, do the enemies evolve to share the same phenotypes like we see in convergent evolution on Earth? Is evolution predictable overall? What might the conclusions from Project Hastur suggest about predictable evolution on Earth? After running trials during which a player used multiple different strategies in many different evolutionary treatments (turning certain fitness functions on/off), we found that the relative amount of variation produced was predictable, but the details of the varieties of enemies that arise differ dramatically even within replicates.

Adding Mate Selection to an Evolutionary Video Game

Evolvy Bugs is an evolutionary mobile video game being developed by the Polymorphic Games Studio at the University of Idaho. Just like other Polymorphic Games projects, the enemies evolve to combat the player’s strategy through a mathematical model of biological evolution. The fitness of enemies is based on how long they survive in the fight against the player, and enemies with the highest fitness reproduce at the end of each generation. Evolvy Bugs takes inspiration from Space Invaders: the player controls a space ship that shoots at aliens that float in from the top of the screen. With the addition of evolution to this classic style of game, players can watch the aliens come up with new behaviors and resistances to combat their strategies. For example, if a player shoots at aliens as soon as they drop down from the top of the screen, the aliens will start to choose to hide behind asteroids to prevent the player from easily shooting at them.

My role in the development of Evolvy Bugs is to add a model of mate selection to the game. In this model of mate selection, females prefer males that have more hair. Due to this selection pressure, players see the alien population get hairier and hairier as the game progresses. The trait under selection does not have a cost, so the aliens display Fisherian selection. This illustration of Fisherian selection adds another evolutionary process that can be taught in classrooms using this game.

If you'd like to hear me talk about this project, I gave an IIDS (Institute for Interdisciplinary Data Sciences) seminar about it in 2021 that was recorded and uploaded to YouTube.

Three males in Evolvy Bugs at different generations. The male on the far-left is from a generation early in gameplay and does not have a lot of hair because mate selection has not impacted the evolution of males at that point. The male in the middle is from about 10 generations later and has more hair due to mate selection. The male on the far-right is from about 20 generations later than the middle male and has the most hair, driven by the females' preference for hairy males.

Gopher Tortoise Commensal Diversity

Inspired by the population of relocated gopher tortoises (Gopherus polyphemus) at Circle B Bar Reserve in Lakeland, FL, I worked to determine whether the commensal diversity among this relocated population of tortoises was different from the commensal diversity among an undisturbed, natural population of gopher tortoises located at Lakeland Highland Scrubs.

Mud Turtle Life History and Spatial Ecology

From 2015 to 2017, I worked on a project that aimed to provide a much-needed update to the literature about the life history and spatial ecology traits of striped mud turtles (Kinosternon baurii). Through mark-recapture techniques, we were able to estimate the population size of mud turtles at Circle B Bar Reserve and make general notes about the overall health of the population. We also conducted a telemetry study that provided insight into the homeranges of the turtles.

Current Research