Research

Modeling symbiosis under stress

Most animals and plants depend on microbial partners, and many of those partnerships are sensitive to warming. Failure can start at any scale: in the flows a host uses to recruit its microbes, in how the microbes adapt, or in how host populations cope. I build mechanistic models at each of these scales, then use sensitivity and identifiability analysis to show experimental collaborators which measurements matter most.

Study systems

A Hawaiian bobtail squid on a dark background

How hosts recruit their microbes

Euprymna scolopes

Hawaiian bobtail squid

Newly hatched squid use breathing and cilia-driven flows to capture their bioluminescent bacterial partner from seawater. I model these flows to ask how heat stress changes that recruitment.

Computational fluid dynamics
Regularized Stokeslets
Sobol sensitivity analysis
Active Brownian motion

Diagram of biofilm formation responding to environmental stressors

How microbes adapt to stress

Vibrio fischeri

Bioluminescent bacterial symbiont

Free-living bacteria must balance surviving in the ocean against colonizing a host. I model how stress selects for biofilm formation, and design the experiments that can measure it.

Trait-structured population models
Natural selection
Identifiability
Genetic algorithms

An Exaiptasia anemone seen from above in a clear container, its long tentacles spread around the mouth

How stressed host populations persist

Exaiptasia diaphana

Sea anemone, a model for coral bleaching

Exaiptasia hosts the same algae as reef-building corals and bleaches the same way. I build population models of how heat stress, and the anemone's microbiome, shape its growth and reproduction.

Delay differential equations
Size-structured models
Generalized additive models
Identifiability

New project

Cassiopea

Upside-down jellyfish

Another animal that lives with photosynthetic algae. With Kaden Muffett, I am using deep learning on daily photographs to measure when Cassiopea polyps transform into juvenile jellyfish.

Deep-learning object detection
Image analysis
Time-series data

Electron micrograph of a Chlamydomonas cell with its two flagella

Active–passive mixtures

Chlamydomonas reinhardtii

Swimming microalga

My PhD work: how swimming algae push passive particles around under confinement, and how to use that to sort them. The physics behind how motile symbionts move near a host's surfaces.

Statistical physics
Active particles
Numerical simulation
Image analysis

Approach

I combine low-Reynolds-number fluid dynamics, population and trait-structured models, and data-driven inference, including machine-learning image analysis. The same toolkit runs through every project: build a model mechanistic enough to explain why a partnership breaks down, then ask which parameters can actually be measured, and which experiments would pin them down, so it can predict when.

All of this work is done in close collaboration with experimental labs: the Heath-Heckman lab (Michigan State), the Nishiguchi lab (UC Merced) and the Sogin lab (UC Riverside), through NSF INSITE, the Institute for Symbiotic Interactions, Training, and Education in the Face of a Changing Climate.

Image credits are given on each project page.