Super-resolution fluorescence microscopy with a resolution beyond the diffraction limit of light, has become an indispensable tool to directly visualize biological structures in living cells at a nanometer-scale resolution. Despite advances in high-density super-resolution fluorescent techniques, existing methods still have bottlenecks, including extremely long execution time, artificial thinning and thickening of structures, and lack of ability to capture latent structures. Here, we propose a novel deep learning guided Bayesian inference (DLBI) approach, for the time-series analysis of high-density fluorescent images. Our method combines the strength of deep learning and statistical inference, where deep learning captures the underlying distribution of the fluorophores that are consistent with the observed time-series fluorescent images by exploring local features and correlation along time-axis, and statistical inference further refines the ultrastructure extracted by deep learning and endues physical meaning to the final image. In particular, our method contains three main components. The first one is a simulator that takes a high-resolution image as the input, and simulates time-series low-resolution fluorescent images based on experimentally calibrated parameters, which provides supervised training data to the deep learning model. The second one is a multi-scale deep learning module to capture both spatial information in each input low-resolution image as well as temporal information among the time-series images. And the third one is a Bayesian inference module that takes the image from the deep learning module as the initial localization of fluorophores and removes artifacts by statistical inference. Comprehensive experimental results on both real and simulated datasets demonstrate that our method provides more accurate and realistic local patch and large-field reconstruction than the state-of-the-art method, the 3B analysis, while our method is more than two orders of magnitude faster. The main program is available at https://github.com/lykaust15/DLBI Supplementary data are available at Bioinformatics online.

The recent developments of far-field optical microscopy (single molecule imaging techniques) have overcome the diffraction barrier of light and improve image resolution by a factor of ten compared with conventional light microscopy. These techniques utilize the stochastic switching of probe molecules to overcome the diffraction limit and determine the precise localizations of molecules, which often requires a long image acquisition time. However, long acquisition times increase the risk of sample drift. In the case of high resolution microscopy, sample drift would decrease the image resolution. In this paper, we propose a novel metric based on the distance between molecules to solve the drift correction. The proposed metric directly uses the position information of molecules to estimate the frame drift. We also designed an algorithm to implement the metric for the general application of drift correction. There are two advantages of our method: First, because our method does not require space binning of positions of molecules but directly operates on the positions, it is more natural for single molecule imaging techniques. Second, our method can estimate drift with a small number of positions in each temporal bin, which may extend its potential application. The effectiveness of our method has been demonstrated by both simulated data and experiments on single molecular images.

Fluorescence nanoscopy has become an indispensable tool for studying organelle structures, protein dynamics, and interactions in biological sciences. Single-molecule localization microscopy can now routinely achieve 10–50 nm resolution through fluorescently labeled specimens in lateral optical sections. However, visualizing structures organized along the axial direction demands scanning and imaging each of the lateral imaging planes with fine intervals throughout the whole cell. This iterative process suffers from photobleaching of tagged probes, is susceptible to alignment artifacts and also limits the imaging speed. Here, we focused on the axial plane super-resolution imaging which integrated the single-objective light-sheet illumination and axial plane optical imaging with single-molecule localization technique to resolve nanoscale cellular architectures along the axial (or depth) dimension without scanning. We demonstrated that this method is compatible with DNA points accumulation for imaging in nanoscale topography (DNA-PAINT) and exchange-PAINT by virtue of its light-sheet illumination, allowing multiplexed super-resolution imaging throughout the depth of whole cells. We further demonstrated this proposed system by resolving the axial distributions of intracellular organelles such as microtubules, mitochondria, and nuclear pore complexes in both COS-7 cells and glioblastoma patient-derived tumor cells.