Through innovations in both imaging techniques and the ability to process these images at scale, high- resolution imaging is transforming the eld of molecular biology, yet its power has yet to be fully utilized for asking questions in evolutionary biology. Just as demographic surveys can reveal more or less densely populated areas where, for example, a contagious disease may spread at di erent rates, these imaging datasets can help us quantify cellular and molecular patterns of spatial variation and understand how this variation a ects rates of evolution, by impeding or accelerating the spread of new variants through the population. My research program, at the interface of computer vision and evolutionary biology, is exploring how molecular and cellular communities spatially organize, and how the resulting spatial topologies can be generated, stably maintained and further shape the outcome of the evolutionary process. What are spatial topologies that act to amplify the selective advantage of new mutations in the pop- ulation, versus structures that dampen the force of selection and slow down rates of evolution? We build theoretical evolutionary models that explore how the rate of evolution is shaped by complex spatial struc- ture and nd the relevant spatial features for evolutionary ampli cation or selective suppression. We link these theoretical population genetic models to high-resolution imaging datasets and study the resulting spatial architectures. This allows us to go beyond simply describing patterns of cellular or molecular spatial variation, and enables exploration of the generative processes, as well as of the evolutionary trajectories of the system. Beyond the purely theoretical interest in these questions, understanding the role of spatial structure in shaping the mode and tempo of evolutionary dynamics is particularly timely because, by using modern microfuidics and organoid technologies, we can start building population structures that control the topol- ogy and migration patterns of a molecular or cellular population, amplifying the selective bene t of chosen mutations, boosting the ability to nd optimized protein complexes for medical or industrial applications, or as a screening tool for faster replicating pathogenic variants.

Through recent innovations in imaging techniques and the ability to process these images at scale, we have unprecedented access to public datasets of molecular and cellular spatial patterns of organization. To make full use of these new spatially-complex data streams, to understand the dynamical processes that are shaping cellular and molecular spatial collectives, to go beyond simple description of pattern towards understanding of function and adaptive design, we must rst develop a predictive theory of complex spatial structure. My group builds mathematical representations of the spatial architecture of molecular and cellular biological systems and studies the evolutionary-relevant properties of these complex spatial topologies.