The complex microbial communities that comprise the human microbiome form intricate spatially structured biofilms on host surfaces, creating networks of interactions and modulating human health. Yet progress in understanding these microscale ecosystems has historically been hindered by limited tools to spatially map the full complexity of microbial communities. We reviewed recent advances in tools to spatially map microbiomes, identifying fluorescence in situ hybridization (FISH) as a particularly promising approach. We then developed an imaging approach that pairs single molecule DNA-FISH with multiplexed ribosomal RNA-FISH, thereby enabling the simultaneous visualization of both mobile genetic elements and their cognate hosts. We revealed that spatial heterogeneity in bacterial taxa results in heterogenous distribution of MGEs within the community, where clusters of MGEs could be local hotspots of horizontal gene transfer or expansion of host strains carrying the MGE. Last, we applied bacterial taxon mapping to study dysbiosis in peri-implant disease. Based on our findings, we propose a model of peri-implant dysbiosis where changes in the spatial structure allow colonization of new community members that provoke the immune system. The methods introduced here can help advance the study of microbial ecology and address pressing questions regarding dysbiosis, antimicrobial resistance, and phage therapy.