Relevant Projects:


Kevin Wood

Associate Professor of Biophysics and Physics

Associate Professor of Biophysics

My group studies a broad collection of problems in biological physics and non-equilibrium statistical physics, but we are particularly focused on spatiotemporal dynamics in microbial communities. We combine quantitative experiments in microbes with mathematical models drawn from statistical physics, dynamical systems, and control theory in an effort to predict, and ultimately control, the evolution of complex cellular communities. For example, we recently showed how tools from reinforcement learning and stochastic control can be leveraged to design multi-drug antibiotic cycles that slow the emergence of resistance in an opportunistic pathogen in the lab (Maltas and Wood, PLOS Biology, 2019). We also use basic machine learning approaches to automate certain features of our data acquisition and analysis pipeline—for example, to segment and identify individual bacterial cells in surface associated (biofilm) communities. We recognize the tremendous potential of data-driven approaches for accelerating research at the intersection of physics and microbiology, and perhaps more importantly, for identifying new theoretical principles to shape the evolution of multi-cellular communities.

Bacterial colonies consisting of drug-sensitive (red) and drug resistant (green) populations exhibit a wide range of spatial patterns as drug concentration (vertical) and resistant fraction (horizontal) varies. The presence of resistant (green) cells in a community allows for survival of neighboring sensitive (red) populations, even at drug concentrations above the MIC (minimum inhibitor concentration) that would otherwise be fatal. Image modified from Sharma and Wood, ISME, 2021.