We embrace an interdisciplinary approach to data science focused on networks and network representations. We use mathematical models and statistical principles to develop and apply computational tools for the study of real-world data, working in close collaboration with domain science experts. With “nodes” representing objects of interest and “edges” that connect the nodes representing relationships or similarities, the concept of a network can be flexibly used across many applications. Most people are familiar with the concept of a network in terms of hyperlinked web pages or online social networks, and online networks are indeed an area of broad interest (including some of our own work), but networks also appear in a much wider variety of connected systems. Example applications in our own work include modeling the spread of different infectious diseases and behaviors, predicting conflict in international relations, assessing the importance of patient care coordination, accurately detecting active C.¬†diff infections from metabolomic data, and understanding various aspects of neural development, disease and degeneracy.

Our research group includes postdoctoral scholars, graduate students, and undergraduate researchers working on different aspects of networks and data science, including developments in community detection, network representations of data, modeling network dynamics, and diffusive processes with applications to disease and health behaviors. Our group is diverse and inclusive, supporting each others’ efforts as we work on individual projects and finding natural collaborations when opportunities present themselves. Because our work is fundamentally interdisciplinary, we also collaborate with a number of other students and faculty from other departments and universities.

All of my accepted and published papers are listed on my CV. Published papers are also listed at Google Scholar. A subset of these, including all public-access-required NIH-sponsored publications, are available through My Bibliography at NLM. Preprints of our work are typically available from or other preprint servers (as indicated on my CV) except where disciplinary norms among my collaborators discourage preprints.

Some old pages:

  • The latest release of our GenLouvain code, a generalized Louvain method for community detection implemented in MATLAB, is available on GitHub.
  • Random Walker Rankings is my old blog with Thomas Callaghan about mathematics and statistics in sports, with special emphasis on our RW/RWFL rankings of college football.
  • NetWiki was our old dual-purpose wiki about network science, including space both for private collaborations and public posting of data and links. Because this was hosted on a virtual machine at my old university, I do not have any easy means to revive it now that it has gone offline in one of their software upgrades. If you are looking for something specific that used to be hosted on that site, you can inquire directly with me.