Single cell segmentation in microbiome imaging

Microbes in nature often live in communities with intricate spatial organization. Recent developments in molecular barcoding strategies and confocal spectral imaging have enabled spatially resolved and highly multiplexed phylogenetic measurements in these communities. However, quantitative analysis of these information-rich imaging dataset remains difficult, primarily due to bottlenecks in accurate single cell segmentation. Here, we present our approach to segment spectral images of environmental microbiome using information contained in the neighborhood of each voxel. We will discuss preliminary segmentation results and quantitative analysis of the spatial organization of microbial communities at the single cell level.