Single-molecule localization and diffusivity microscopy reveals dynamic biomolecular organization in living cells.

Single-molecule tracking in living cells measures protein diffusivity but requires sparse imaging, limiting high-density mapping. Here we introduce single-molecule localization and diffusivity microscopy (SMLDM), a deep learning-based approach that accurately estimates single-molecule movement tracks and diffusion coefficients directly from single-frame snapshots, eliminating the need for trajectory linking. Implemented as mobility photoactivated localization microscopy (MPALM) with bright photoactivatable fluorophores and U-Net-based single-molecule segmentation, this method achieves a 50- to 300-fold increase in data density compared to conventional tracking-based approaches, generating high-density, spatially super-resolved maps of molecular diffusivity and organization in living human cells. We applied MPALM to diverse dynamic cellular processes, uncovering nucleosome clustering into low-mobility chromatin domains, pathway-biased μ-opioid receptor dynamic clustering, focal adhesion movement and nonuniform molecular diffusivity and microcondensate organization during early droplet coalescence. SMLDM provides a powerful tool for resolving biomolecular organization and dynamics at single-molecule resolution in live cells.