Project Summary Bacterial biofilms are surface-attached communities of bacterial cells enclosed in an extracellular matrix. Biofilms are a concern in health and in industrial operations because of persistent infections, clogging of flows, and surface fouling. Recent advances in single-cell live imaging have revealed well-defined cell ordering in individual biofilm clusters and the underlying biomechanical principles that shape them. However, we have little understanding of which genes are activated in each biofilm-dwelling cell and how cell organization is determined by the gene expression pattern at the single-cell level. We do not know whether and how gene expression profiles vary from cell to cell in biofilms, and what consequences such heterogeneity has on biofilm development. Cell-to-cell variation in biofilms could underly the notorious difficulty in eradicating biofilms in chronic infections because of the differential response of biofilm cells to antibiotic treatment. In this proposal we put forward ideas to address this challenge by developing new imaging platforms to capture the biofilm growth dynamics and associated gene expression pattern at the single-cell level. Using these imaging platforms, we will uncover the intricate interplay between single-cell gene expression, individual cell behavior, and local cell organization that underlies the developmental program of bacterial biofilms. Specifically, we will use deep learning algorithms to push the temporal and spatial resolution limits in single-cell biofilm imaging, and develop an innovative, coupled segmentation-tracking method to generate a robust three-dimensional lineage tracing algorithm in growing biofilms. By combining lineage tracing and fluorescent reporters, we will follow the spatiotemporal expression pattern of each individual cell throughout biofilm development. By focusing on matrix production, degradation, and cell dispersion, we will create concrete examples of gene expression heterogeneity at the single-cell level and the associated consequences in 3D biofilms. In addition, we will investigate heterogeneity in genes involved in intra- and intercellular signaling, in motility and attachment, and in cell shape regulation to broaden our finding. With these efforts, we will reveal how individual gene expression, cell behavior, and local cell ordering reciprocally interact with each other to define biofilm architecture and development. The knowledge obtained in the current proposal offers new strategies for manipulating complex bacterial communities and has the potential to change the clinical procedure of treating biofilm-related diseases, for example by developing new chemicals that specifically induce dispersal or target the recalcitrant cell populations.