A little bit about me

The tough part about being a mathematical biologist is that the mathematicians think you've gone soft and the biologists think you've gone crazy.

I am a mathematical biologist. I love developing novel mathematical tools motivated by biological problems, and I love using mathematics to interrogate the logical consequences of biological theories for more informative study design.

My main interests are in community ecology, community assembly, evolution/co-evolution, and stochastic processes. I'm also interested in human systems, such as financial markets and human memetic communities (the ideas floating around in our heads) as evolving systems with data complementary to what we can collect in biology.

My side interests are outdoor sports, philosophy, trading strategy development and portfolio management, and sipping whiskey by a fireplace.


Phylofactorization - How can we simplify microbiome data using phylogenies? My research on phylogenetic factorization developed a new way to simplify biological big-data by constructing variables corresponding to edges in the tree of life. Above, we've characterized microbes in the American Gut project with a phylogenetic analog of principal components analysis, yielding phylogenetic components of variance in the gut microbiome.

Mapping our inferences onto the tree of life enables further studies of microbial physiology and genome biology to understand why microbes in different parts of the evolutionary tree have different habitat preferences.

Diversity Dynamics - How does the diversity of a community change over time? Many stochastic community models look at population dynamics, but relative abundances carry important information about competitive inequalities.


Diversity/evenness  metrics tell us how close a community is to the edges of the simplex where extinctions occur. Understanding the dynamics of diversity can provide a novel mathematical approach to understanding trophic island biogeography and hypothesis tests of diversity metrics - is diversity higher/lower or more/less volatile than our null model - can allow us to estimate the risk of extinctions.