Super-resolution fight club: assessment of 2D and 3D single-molecule localization microscopy software

Build and operation of a custom 3D, multicolor, single-molecule localization microscope

Single-molecule localization microscopy (SMLM) enables imaging scientists to visualize biological structures with unprecedented resolution. Particularly powerful implementations of SMLM are capable of three-dimensional, multicolor and high-throughput imaging and can yield key biological insights. However, widespread access to these technologies is limited, primarily by the cost of commercial options and complexity of de novo development of custom systems. Here we provide a comprehensive guide for interested researchers who wish to establish a high-end, custom-built SMLM setup in their laboratories. We detail the initial configuration and subsequent assembly of the SMLM, including the instructions for the alignment of all the optical pathways, the software and hardware integration, and the operation of the instrument. We describe the validation steps, including the preparation and imaging of test and biological samples with structures of well-defined geometries, and assist the user in troubleshooting and benchmarking the system's performance. Additionally, we provide a walkthrough of the reconstruction of a super-resolved dataset from acquired raw images using the Super-resolution Microscopy Analysis Platform. Depending on the instrument configuration, the cost of the components is in the range US$95,000-180,000, similar to other open-source advanced SMLMs, and substantially lower than the cost of a commercial instrument. A builder with some experience of optical systems is expected to require 4-8 months from the start of the system construction to attain high-quality three-dimensional and multicolor biological images.

Super-resolution microscopies, technological breakthrough to decipher mitochondrial structure and dynamic

Mitochondria are complex organelles with an outer membrane enveloping a second inner membrane that creates a vast matrix space partitioned by pockets or cristae that join the peripheral inner membrane with several thin junctions. Several micrometres long, mitochondria are generally close to 300 nm in diameter, with membrane layers separated by a few tens of nanometres. Ultrastructural data from electron microscopy revealed the structure of these mitochondria, while conventional optical microscopy revealed their extraordinary dynamics through fusion, fission, and migration processes but its limited resolution power restricted the possibility to go further. By overcoming the limits of light diffraction, Super-Resolution Microscopy (SRM) now offers the potential to establish the links between the ultrastructure and remodelling of mitochondrial membranes, leading to major advances in our understanding of mitochondria's structure-function. Here we review the contributions of SRM imaging to our understanding of the relationship between mitochondrial structure and function. What are the hopes for these new imaging approaches which are particularly important for mitochondrial pathologies?

Deep learning-based spectroscopic single-molecule localization microscopy

Significance

Spectroscopic single-molecule localization microscopy (sSMLM) takes advantage of nanoscopy and spectroscopy, enabling sub-10 nm resolution as well as simultaneous multicolor imaging of multi-labeled samples. Reconstruction of raw sSMLM data using deep learning is a promising approach for visualizing the subcellular structures at the nanoscale.

Aim

Develop a novel computational approach leveraging deep learning to reconstruct both label-free and fluorescence-labeled sSMLM imaging data.

Approach

We developed a two-network-model based deep learning algorithm, termed DsSMLM, to reconstruct sSMLM data. The effectiveness of DsSMLM was assessed by conducting imaging experiments on diverse samples, including label-free single-stranded DNA (ssDNA) fiber, fluorescence-labeled histone markers on COS-7 and U2OS cells, and simultaneous multicolor imaging of synthetic DNA origami nanoruler.

Results

For label-free imaging, a spatial resolution of 6.22 nm was achieved on ssDNA fiber; for fluorescence-labeled imaging, DsSMLM revealed the distribution of chromatin-rich and chromatin-poor regions defined by histone markers on the cell nucleus and also offered simultaneous multicolor imaging of nanoruler samples, distinguishing two dyes labeled in three emitting points with a separation distance of 40 nm. With DsSMLM, we observed enhanced spectral profiles with 8.8% higher localization detection for single-color imaging and up to 5.05% higher localization detection for simultaneous two-color imaging.

Conclusions

We demonstrate the feasibility of deep learning-based reconstruction for sSMLM imaging applicable to label-free and fluorescence-labeled sSMLM imaging data. We anticipate our technique will be a valuable tool for high-quality super-resolution imaging for a deeper understanding of DNA molecules' photophysics and will facilitate the investigation of multiple nanoscopic cellular structures and their interactions.

OneFlowTraX: a user-friendly software for super-resolution analysis of single-molecule dynamics and nanoscale organization

Super-resolution microscopy (SRM) approaches revolutionize cell biology by providing insights into the nanoscale organization and dynamics of macromolecular assemblies and single molecules in living cells. A major hurdle limiting SRM democratization is post-acquisition data analysis which is often complex and time-consuming. Here, we present OneFlowTraX, a user-friendly and open-source software dedicated to the analysis of single-molecule localization microscopy (SMLM) approaches such as single-particle tracking photoactivated localization microscopy (sptPALM). Through an intuitive graphical user interface, OneFlowTraX provides an automated all-in-one solution for single-molecule localization, tracking, as well as mobility and clustering analyses. OneFlowTraX allows the extraction of diffusion and clustering parameters of millions of molecules in a few minutes. Finally, OneFlowTraX greatly simplifies data management following the FAIR (Findable, Accessible, Interoperable, Reusable) principles. We provide a detailed step-by-step manual and guidelines to assess the quality of single-molecule analyses. Applying different fluorophores including mEos3.2, PA-GFP, and PATagRFP, we exemplarily used OneFlowTraX to analyze the dynamics of plant plasma membrane-localized proteins including an aquaporin, the brassinosteroid receptor Brassinosteroid Insensitive 1 (BRI1) and the Receptor-Like Protein 44 (RLP44).

RASP: Optimal Single Puncta Detection in Complex Cellular Backgrounds

Super-resolution and single-molecule microscopies have been increasingly applied to complex biological systems. A major challenge of these approaches is that fluorescent puncta must be detected in the low signal, high noise, heterogeneous background environments of cells and tissue. We present RASP, Radiality Analysis of Single Puncta, a bioimaging-segmentation method that solves this problem. RASP removes false-positive puncta that other analysis methods detect and detects features over a broad range of spatial scales: from single proteins to complex cell phenotypes. RASP outperforms the state-of-the-art methods in precision and speed using image gradients to separate Gaussian-shaped objects from the background. We demonstrate RASP's power by showing that it can extract spatial correlations between microglia, neurons, and α-synuclein oligomers in the human brain. This sensitive, computationally efficient approach enables fluorescent puncta and cellular features to be distinguished in cellular and tissue environments, with sensitivity down to the level of the single protein. Python and MATLAB codes, enabling users to perform this RASP analysis on their own data, are provided as Supporting Information and links to third-party repositories.

Photophysical image analysis: Unsupervised probabilistic thresholding for images from electron-multiplying charge-coupled devices

We introduce the concept photophysical image analysis (PIA) and an associated pipeline for unsupervised probabilistic image thresholding for images recorded by electron-multiplying charge-coupled device (EMCCD) cameras. We base our approach on a closed-form analytic expression for the characteristic function (Fourier-transform of the probability mass function) for the image counts recorded in an EMCCD camera, which takes into account both stochasticity in the arrival of photons at the imaging camera and subsequent noise induced by the detection system of the camera. The only assumption in our method is that the background photon arrival to the imaging system is described by a stationary Poisson process (we make no assumption about the photon statistics for the signal). We estimate the background photon statistics parameter, λbg, from an image which contains both background and signal pixels by use of a novel truncated fit procedure with an automatically determined image count threshold. Prior to this, the camera noise model parameters are estimated using a calibration step. Utilizing the estimates for the camera parameters and λbg, we then introduce a probabilistic thresholding method, where, for the first time, the fraction of misclassified pixels can be determined a priori for a general image in an unsupervised way. We use synthetic images to validate our a priori estimates and to benchmark against the Otsu method, which is a popular unsupervised non-probabilistic image thresholding method (no a priori estimates for the error rates are provided). For completeness, we lastly present a simple heuristic general-purpose segmentation method based on the thresholding results, which we apply to segmentation of synthetic images and experimental images of fluorescent beads and lung cell nuclei. Our publicly available software opens up for fully automated, unsupervised, probabilistic photophysical image analysis.

High-density volumetric super-resolution microscopy

Volumetric super-resolution microscopy typically encodes the 3D position of single-molecule fluorescence into a 2D image by changing the shape of the point spread function (PSF) as a function of depth. However, the resulting large and complex PSF spatial footprints reduce biological throughput and applicability by requiring lower labeling densities to avoid overlapping fluorescent signals. We quantitatively compare the density dependence of single-molecule light field microscopy (SMLFM) to other 3D PSFs (astigmatism, double helix and tetrapod) showing that SMLFM enables an order-of-magnitude speed improvement compared to the double helix PSF by resolving overlapping emitters through parallax. We demonstrate this optical robustness experimentally with high accuracy ( > 99.2 ± 0.1%, 0.1 locs μm-2) and sensitivity ( > 86.6 ± 0.9%, 0.1 locs μm-2) through whole-cell (scan-free) imaging and tracking of single membrane proteins in live primary B cells. We also exemplify high-density volumetric imaging (0.15 locs μm-2) in dense cytosolic tubulin datasets.

NanoLocz: Image Analysis Platform for AFM, High-Speed AFM, and Localization AFM

Atomic Force Microscopy (AFM), High-Speed AFM (HS-AFM) simulation AFM, and Localization AFM (LAFM) enable the study of molecules and surfaces with increasingly higher spatiotemporal resolution. However, effective and rapid analysis of the images and movies produced by these techniques can be challenging, often requiring the use of multiple image processing software applications and scripts. Here, NanoLocz, an open-source solution that offers advanced analysis capabilities for the AFM community, is presented. Integration and continued development of AFM analysis tools is essential to improve access to data, increase throughput, and open new analysis opportunities. NanoLocz efficiently leverages the rich data AFM has to offer by incorporating and combining existing and newly developed analysis methods for AFM, HS-AFM, simulation AFM, and LAFM seamlessly. It facilitates and streamlines AFM analysis workflows from import of raw data, through to various analysis workflows. Here, the study demonstrates the capabilities of NanoLocz and the new methods it enables including single-molecule LAFM, time-resolved LAFM, and simulation LAFM.

Antibody agonists trigger immune receptor signaling through local exclusion of receptor-type protein tyrosine phosphatases

Antibodies can block immune receptor engagement or trigger the receptor machinery to initiate signaling. We hypothesized that antibody agonists trigger signaling by sterically excluding large receptor-type protein tyrosine phosphatases (RPTPs) such as CD45 from sites of receptor engagement. An agonist targeting the costimulatory receptor CD28 produced signals that depended on antibody immobilization and were sensitive to the sizes of the receptor, the RPTPs, and the antibody itself. Although both the agonist and a non-agonistic anti-CD28 antibody locally excluded CD45, the agonistic antibody was more effective. An anti-PD-1 antibody that bound membrane proximally excluded CD45, triggered Src homology 2 domain-containing phosphatase 2 recruitment, and suppressed systemic lupus erythematosus and delayed-type hypersensitivity in experimental models. Paradoxically, nivolumab and pembrolizumab, anti-PD-1-blocking antibodies used clinically, also excluded CD45 and were agonistic in certain settings. Reducing these agonistic effects using antibody engineering improved PD-1 blockade. These findings establish a framework for developing new and improved therapies for autoimmunity and cancer.

Harnessing artificial intelligence to reduce phototoxicity in live imaging

Fluorescence microscopy is essential for studying living cells, tissues and organisms. However, the fluorescent light that switches on fluorescent molecules also harms the samples, jeopardizing the validity of results - particularly in techniques such as super-resolution microscopy, which demands extended illumination. Artificial intelligence (AI)-enabled software capable of denoising, image restoration, temporal interpolation or cross-modal style transfer has great potential to rescue live imaging data and limit photodamage. Yet we believe the focus should be on maintaining light-induced damage at levels that preserve natural cell behaviour. In this Opinion piece, we argue that a shift in role for AIs is needed - AI should be used to extract rich insights from gentle imaging rather than recover compromised data from harsh illumination. Although AI can enhance imaging, our ultimate goal should be to uncover biological truths, not just retrieve data. It is essential to prioritize minimizing photodamage over merely pushing technical limits. Our approach is aimed towards gentle acquisition and observation of undisturbed living systems, aligning with the essence of live-cell fluorescence microscopy.

Mapping super-resolution image quality

The local quality of super-resolution microscopy images can be assessed and mapped by rolling Fourier ring correlation, even when image quality varies within a single image.

In Situ Imaging of Proteins Using DNA-PAINT Super-Resolution Microscopy

The spatial resolution of conventional light microscopy is restricted by the diffraction limit to hundreds of nanometers. Super-resolution microscopy enables single digit nanometer resolution by circumventing the diffraction limit of conventional light microscopy. DNA point accumulation for imaging in nanoscale topography (DNA-PAINT) belongs to the family of single-molecule localization super-resolution approaches. Unique features of DNA-PAINT are that it allows for sub-nanometer resolution, spectrally unlimited multiplexing, proximity detection, and quantitative counting of target molecules. Here, we describe prerequisites for efficient DNA-PAINT microscopy.

Advanced optical imaging for the rational design of nanomedicines

Despite the enormous potential of nanomedicines to shape the future of medicine, their clinical translation remains suboptimal. Translational challenges are present in every step of the development pipeline, from a lack of understanding of patient heterogeneity to insufficient insights on nanoparticle properties and their impact on material-cell interactions. Here, we discuss how the adoption of advanced optical microscopy techniques, such as super-resolution optical microscopies, correlative techniques, and high-content modalities, could aid the rational design of nanocarriers, by characterizing the cell, the nanomaterial, and their interaction with unprecedented spatial and/or temporal detail. In this nanomedicine arena, we will discuss how the implementation of these techniques, with their versatility and specificity, can yield high volumes of multi-parametric data; and how machine learning can aid the rapid advances in microscopy: from image acquisition to data interpretation.

Physically informed Monte Carlo simulation of dual-wedge prism-based spectroscopic single-molecule localization microscopy

Significance

The dual-wedge prism (DWP)-based spectroscopic single-molecule localization microscopy (sSMLM) system offers improved localization precision and adjustable spectral or localization performance, but its nonlinear spectral dispersion presents a challenge. A systematic method can help understand the challenges and thereafter optimize the DWP system's performance by customizing the system parameters to maximize the spectral or localization performance for various molecular labels.

Aim

We developed a Monte Carlo (MC)-based model that predicts the imaging output of the DWP-based sSMLM system given different system parameters.

Approach

We assessed our MC model's localization and spectral precisions by comparing our simulation against theoretical equations and fluorescent microspheres. Furthermore, we simulated the DWP-based system using beamsplitters (BSs) with a reflectance (R):transmittance (T) of R50:T50 and R30:T70 and their tradeoffs.

Results

Our MC simulation showed average deviations of 2.5 and 2.1 nm for localization and spectral precisions against theoretical equations and 2.3 and 1.0 nm against fluorescent microspheres. An R30:T70 BS improved the spectral precision by 8% but worsened the localization precision by 35% on average compared with an R50:T50 BS.

Conclusions

The MC model accurately predicted the localization precision, spectral precision, spectral peaks, and spectral widths of fluorescent microspheres, as validated by experimental data. Our work enhances the theoretical understanding of DWP-based sSMLM for multiplexed imaging, enabling performance optimization.

Visualizing HIV-1 Assembly at the T-Cell Plasma Membrane Using Single-Molecule Localization Microscopy

The 20-year revolution in optical fluorescence microscopy, supported by the optimization of both spatial resolution and timely acquisition, allows the visualization of nanoscaled objects in cell biology. Currently, the use of a recent generation of super-resolution fluorescence microscope coupled with improved fluorescent probes gives the possibility to study the replicative cycle of viruses in living cells, at the single-virus particle or protein level. Here, we highlight the protocol for visualizing HIV-1 Gag assembly at the host T-cell plasma membrane using super-resolution light microscopy. Total internal reflection fluorescence microscopy (TIRF-M) coupled with single-molecule localization microscopy (SMLM) enables the detection and characterization of the assembly of viral proteins at the plasma membrane of infected host cells at the single protein level. Here, we describe the TIRF equipment, the T-cell culture for HIV-1, the sample preparation for single-molecule localization microscopies such as PALM and STORM, acquisition protocols, and Gag assembling cluster analysis.

Quantitatively mapping local quality of super-resolution microscopy by rolling Fourier ring correlation

In fluorescence microscopy, computational algorithms have been developed to suppress noise, enhance contrast, and even enable super-resolution (SR). However, the local quality of the images may vary on multiple scales, and these differences can lead to misconceptions. Current mapping methods fail to finely estimate the local quality, challenging to associate the SR scale content. Here, we develop a rolling Fourier ring correlation (rFRC) method to evaluate the reconstruction uncertainties down to SR scale. To visually pinpoint regions with low reliability, a filtered rFRC is combined with a modified resolution-scaled error map (RSM), offering a comprehensive and concise map for further examination. We demonstrate their performances on various SR imaging modalities, and the resulting quantitative maps enable better SR images integrated from different reconstructions. Overall, we expect that our framework can become a routinely used tool for biologists in assessing their image datasets in general and inspire further advances in the rapidly developing field of computational imaging.

Spatial redundancy transformer for self-supervised fluorescence image denoising

Fluorescence imaging with high signal-to-noise ratios has become the foundation of accurate visualization and analysis of biological phenomena. However, the inevitable noise poses a formidable challenge to imaging sensitivity. Here we provide the spatial redundancy denoising transformer (SRDTrans) to remove noise from fluorescence images in a self-supervised manner. First, a sampling strategy based on spatial redundancy is proposed to extract adjacent orthogonal training pairs, which eliminates the dependence on high imaging speed. Second, we designed a lightweight spatiotemporal transformer architecture to capture long-range dependencies and high-resolution features at low computational cost. SRDTrans can restore high-frequency information without producing oversmoothed structures and distorted fluorescence traces. Finally, we demonstrate the state-of-the-art denoising performance of SRDTrans on single-molecule localization microscopy and two-photon volumetric calcium imaging. SRDTrans does not contain any assumptions about the imaging process and the sample, thus can be easily extended to various imaging modalities and biological applications.

High-fidelity 3D live-cell nanoscopy through data-driven enhanced super-resolution radial fluctuation

Live-cell super-resolution microscopy enables the imaging of biological structure dynamics below the diffraction limit. Here we present enhanced super-resolution radial fluctuations (eSRRF), substantially improving image fidelity and resolution compared to the original SRRF method. eSRRF incorporates automated parameter optimization based on the data itself, giving insight into the trade-off between resolution and fidelity. We demonstrate eSRRF across a range of imaging modalities and biological systems. Notably, we extend eSRRF to three dimensions by combining it with multifocus microscopy. This realizes live-cell volumetric super-resolution imaging with an acquisition speed of ~1 volume per second. eSRRF provides an accessible super-resolution approach, maximizing information extraction across varied experimental conditions while minimizing artifacts. Its optimal parameter prediction strategy is generalizable, moving toward unbiased and optimized analyses in super-resolution microscopy.

A generative model to simulate spatiotemporal dynamics of biomolecules in cells

Generators of space-time dynamics in bioimaging have become essential to build ground truth datasets for image processing algorithm evaluation such as biomolecule detectors and trackers, as well as to generate training datasets for deep learning algorithms. In this contribution, we leverage a stochastic model, called birth-death-move (BDM) point process, in order to generate joint dynamics of biomolecules in cells. This particle-based stochastic simulation method is very flexible and can be seen as a generalization of well-established standard particle-based generators. In comparison, our approach allows us: (1) to model a system of particles in motion, possibly in interaction, that can each possibly switch from a motion regime (e.g., Brownian) to another (e.g., a directed motion); (2) to take into account finely the appearance over time of new trajectories and their disappearance, these events possibly depending on the cell regions but also on the current spatial configuration of all existing particles. This flexibility enables to generate more realistic dynamics than standard particle-based simulation procedures, by for example accounting for the colocalization phenomena often observed between intracellular vesicles. We explain how to specify all characteristics of a BDM model, with many practical examples that are relevant for bioimaging applications. As an illustration, based on real fluorescence microscopy datasets, we finally calibrate our model to mimic the joint dynamics of Langerin and Rab11 proteins near the plasma membrane, including the well-known colocalization occurrence between these two types of vesicles. We show that the resulting synthetic sequences exhibit comparable features as those observed in real microscopy image sequences.

Resolution enhancement with deblurring by pixel reassignment

Improving the spatial resolution of a fluorescence microscope has been an ongoing challenge in the imaging community. To address this challenge, a variety of approaches have been taken, ranging from instrumentation development to image postprocessing. An example of the latter is deconvolution, where images are numerically deblurred based on a knowledge of the microscope point spread function. However, deconvolution can easily lead to noise-amplification artifacts. Deblurring by postprocessing can also lead to negativities or fail to conserve local linearity between sample and image. We describe here a simple image deblurring algorithm based on pixel reassignment that inherently avoids such artifacts and can be applied to general microscope modalities and fluorophore types. Our algorithm helps distinguish nearby fluorophores, even when these are separated by distances smaller than the conventional resolution limit, helping facilitate, for example, the application of single-molecule localization microscopy in dense samples. We demonstrate the versatility and performance of our algorithm under a variety of imaging conditions.

Simulating structurally variable nuclear pore complexes for microscopy

MOTIVATION

The nuclear pore complex (NPC) is the only passageway for macromolecules between nucleus and cytoplasm, and an important reference standard in microscopy: it is massive and stereotypically arranged. The average architecture of NPC proteins has been resolved with pseudoatomic precision, however observed NPC heterogeneities evidence a high degree of divergence from this average. Single-molecule localization microscopy (SMLM) images NPCs at protein-level resolution, whereupon image analysis software studies NPC variability. However, the true picture of this variability is unknown. In quantitative image analysis experiments, it is thus difficult to distinguish intrinsically high SMLM noise from variability of the underlying structure.

RESULTS

We introduce CIR4MICS ('ceramics', Configurable, Irregular Rings FOR MICroscopy Simulations), a pipeline that synthesizes ground truth datasets of structurally variable NPCs based on architectural models of the true NPC. Users can select one or more N- or C-terminally tagged NPC proteins, and simulate a wide range of geometric variations. We also represent the NPC as a spring-model such that arbitrary deforming forces, of user-defined magnitudes, simulate irregularly shaped variations. Further, we provide annotated reference datasets of simulated human NPCs, which facilitate a side-by-side comparison with real data. To demonstrate, we synthetically replicate a geometric analysis of real NPC radii and reveal that a range of simulated variability parameters can lead to observed results. Our simulator is therefore valuable to test the capabilities of image analysis methods, as well as to inform experimentalists about the requirements of hypothesis-driven imaging studies.

AVAILABILITY AND IMPLEMENTATION

Code: https://github.com/uhlmanngroup/cir4mics. Simulated data: BioStudies S-BSST1058.

Using single molecule imaging to explore intracellular heterogeneity

Despite more than 100 years of study, it is unclear if the movement of proteins inside the cell is best described as a mosh pit or an exquisitely choreographed dance. Recent studies suggest the latter. Local interactions induce molecular condensates such as liquid-liquid phase separations (LLPSs) or non-liquid, functionally significant molecular aggregates, including synaptic densities, nucleoli, and Amyloid fibrils. Molecular condensates trigger intracellular signaling and drive processes ranging from gene expression to cell division. However, the descriptions of condensates tend to be qualitative and correlative. Here, we indicate how single-molecule imaging and analyses can be applied to quantify condensates. We discuss the pros and cons of different techniques for measuring differences between transient molecular behaviors inside and outside condensates. Finally, we offer suggestions for how imaging and analyses from different time and space regimes can be combined to identify molecular behaviors indicative of condensates within the dynamic high-density intracellular environment.

Spontaneously Blinking Rhodamine Dyes for Single-Molecule Localization Microscopy

Single-molecule localization microscopy (SMLM) has found extensive applications in various fields of biology and chemistry. As a vital component of SMLM, fluorophores play an essential role in obtaining super-resolution fluorescence images. Recent research on spontaneously blinking fluorophores has greatly simplified the experimental setups and extended the imaging duration of SMLM. To support this crucial development, this review provides a comprehensive overview of the development of spontaneously blinking rhodamines from 2014 to 2023, as well as the key mechanistic aspects of intramolecular spirocyclization reactions. We hope that by offering insightful design guidelines, this review will contribute to accelerating the advancement of super-resolution imaging technologies.

Visualizing DNA damage and repair using single molecule super resolution microscopy

Single molecule super resolution microscopy overcomes the diffraction limit by separating individual fluorophore emissions over time, resulting in spatial resolutions that are far superior to epifluorescence microscopy. This allows for DNA damage response (DDR) events to be investigated in greater detail. A variety of DNA damaging drugs can be used on S-phase synchronized immortalized cell lines alongside 5-ethynyl-2'-deoxyuridine (EdU) pulse labelling to ultimately visualize DNA repair pathways at distinct time points and quantify colocalizations between nascent DNA and immunolabeled DDR proteins. This chapter will outline super resolution microscopy assays to interrogate the spatiotemporal organization of DNA repair proteins at damaged foci during DDR events within immortalized cell lines.

Using Single Molecule Imaging to Explore Intracellular Heterogeneity

Despite more than 100 years of study, it is unclear if the movement of proteins inside the cell is best described as a mosh pit or an exquisitely choreographed dance. Recent studies suggest the latter. Local interactions induce molecular condensates such as liquid-liquid phase separations (LLPSs) or non-liquid, functionally significant molecular aggregates, including synaptic densities, nucleoli, and Amyloid fibrils. Molecular condensates trigger intracellular signaling and drive processes ranging from gene expression to cell division. However, the descriptions of condensates tend to be qualitative and correlative. Here, we indicate how single-molecule imaging and analyses can be applied to quantify condensates. We discuss the pros and cons of different techniques for measuring differences between transient molecular behaviors inside and outside condensates. Finally, we offer suggestions for how imaging and analyses from different time and space regimes can be combined to identify molecular behaviors indicative of condensates within the dynamic high-density intracellular environment.

RegiSTORM: channel registration for multi-color stochastic optical reconstruction microscopy

BACKGROUND

Stochastic optical reconstruction microscopy (STORM), a super-resolution microscopy technique based on single-molecule localizations, has become popular to characterize sub-diffraction limit targets. However, due to lengthy image acquisition, STORM recordings are prone to sample drift. Existing cross-correlation or fiducial marker-based algorithms allow correcting the drift within each channel, but misalignment between channels remains due to interchannel drift accumulating during sequential channel acquisition. This is a major drawback in multi-color STORM, a technique of utmost importance for the characterization of various biological interactions.

RESULTS

We developed RegiSTORM, a software for reducing channel misalignment by accurately registering STORM channels utilizing fiducial markers in the sample. RegiSTORM identifies fiducials from the STORM localization data based on their non-blinking nature and uses them as landmarks for channel registration. We first demonstrated accurate registration on recordings of fiducials only, as evidenced by significantly reduced target registration error with all the tested channel combinations. Next, we validated the performance in a more practically relevant setup on cells multi-stained for tubulin. Finally, we showed that RegiSTORM successfully registers two-color STORM recordings of cargo-loaded lipid nanoparticles without fiducials, demonstrating the broader applicability of this software.

CONCLUSIONS

The developed RegiSTORM software was demonstrated to be able to accurately register multiple STORM channels and is freely available as open-source (MIT license) at https://github.com/oystein676/RegiSTORM.git and https://doi.org/10.5281/zenodo.5509861 (archived), and runs as a standalone executable (Windows) or via Python (Mac OS, Linux).

Fluorescence-Based Detection of Proteins and Their Interactions in Live Cells

Recent advances in fluorescence-based microscopy techniques, such as single molecule fluorescence, Förster resonance energy transfer (FRET), fluorescence intensity fluctuations analysis, and super-resolution microscopy have expanded our ability to study proteins in greater detail within their native cellular environment and to investigate the roles that protein interactions play in biological functions, such as inter- and intracellular signaling and cargo transport. In this Perspective, we provide an up-to-date overview of the current state of the art in fluorescence-based detection of proteins and their interactions in living cells with an emphasis on recent developments that have facilitated the characterization of the spatial and temporal organization of proteins into oligomeric complexes in the presence and absence of natural and artificial ligands. Further advancements in this field will only deepen our understanding of the underlying mechanisms of biological processes and help develop new therapeutic targets.

Simple, Economic, and Robust Rail-Based Setup for Super-Resolution Localization Microscopy

Research during the past 2 decades has showcased the power of single-molecule localization microscopy (SMLM) as a tool for exploring the nanoworld. However, SMLM systems are typically available in specialized laboratories and imaging facilities, owing to their expensiveness as well as complex assembly and alignment procedure. Here, we lay out the blueprint of a sturdy, rail-based, cost-efficient, multicolor SMLM setup that is easy to construct and align in service of simplifying the accessibility of SMLM. We characterize the optical properties of the design and assess its capabilities, robustness, and stability. The performance of the system is assayed using super-resolution imaging of biological samples. We believe that this design will make SMLM more affordable and broaden its availability.

Quantitative Single-Molecule Localization Microscopy

Super-resolution fluorescence microscopy allows the investigation of cellular structures at nanoscale resolution using light. Current developments in super-resolution microscopy have focused on reliable quantification of the underlying biological data. In this review, we first describe the basic principles of super-resolution microscopy techniques such as stimulated emission depletion (STED) microscopy and single-molecule localization microscopy (SMLM), and then give a broad overview of methodological developments to quantify super-resolution data, particularly those geared toward SMLM data. We cover commonly used techniques such as spatial point pattern analysis, colocalization, and protein copy number quantification but also describe more advanced techniques such as structural modeling, single-particle tracking, and biosensing. Finally, we provide an outlook on exciting new research directions to which quantitative super-resolution microscopy might be applied.

Purification Analysis, Intracellular Tracking, and Colocalization of Extracellular Vesicles Using Atomic Force and 3D Single-Molecule Localization Microscopy

Extracellular vesicles (EVs) play a key role in cell-cell communication and thus have great potential to be utilized as therapeutic agents and diagnostic tools. In this study, we implemented single-molecule microscopy techniques as a toolbox for a comprehensive characterization as well as measurement of the cellular uptake of HEK293T cell-derived EVs (eGFP-labeled) in HeLa cells. A combination of fluorescence and atomic force microscopy revealed a fraction of 68% fluorescently labeled EVs with an average size of ∼45 nm. Two-color single-molecule fluorescence microscopy analysis elucidated the 3D dynamics of EVs entering HeLa cells. 3D colocalization analysis of two-color direct stochastic optical reconstruction microscopy (dSTORM) images revealed that 25% of EVs that experienced uptake colocalized with transferrin, which has been linked to early recycling of endosomes and clathrin-mediated endocytosis. The localization analysis was combined with stepwise photobleaching, providing a comparison of protein aggregation outside and inside the cells.

Antigen discrimination by T cells relies on size-constrained microvillar contact

T cells use finger-like protrusions called 'microvilli' to interrogate their targets, but why they do so is unknown. To form contacts, T cells must overcome the highly charged, barrier-like layer of large molecules forming a target cell's glycocalyx. Here, T cells are observed to use microvilli to breach a model glycocalyx barrier, forming numerous small (<0.5 μm diameter) contacts each of which is stabilized by the small adhesive protein CD2 expressed by the T cell, and excludes large proteins including CD45, allowing sensitive, antigen dependent TCR signaling. In the absence of the glycocalyx or when microvillar contact-size is increased by enhancing CD2 expression, strong signaling occurs that is no longer antigen dependent. Our observations suggest that, modulated by the opposing effects of the target cell glycocalyx and small adhesive proteins, the use of microvilli equips T cells with the ability to effect discriminatory receptor signaling.

Field-dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging

Single-molecule localization microscopy in a typical wide-field setup has been widely used for investigating subcellular structures with super resolution; however, field-dependent aberrations restrict the field of view (FOV) to only tens of micrometers. Here, we present a deep-learning method for precise localization of spatially variant point emitters (FD-DeepLoc) over a large FOV covering the full chip of a modern sCMOS camera. Using a graphic processing unit-based vectorial point spread function (PSF) fitter, we can fast and accurately model the spatially variant PSF of a high numerical aperture objective in the entire FOV. Combined with deformable mirror-based optimal PSF engineering, we demonstrate high-accuracy three-dimensional single-molecule localization microscopy over a volume of ~180 × 180 × 5 μm³, allowing us to image mitochondria and nuclear pore complexes in entire cells in a single imaging cycle without hardware scanning; a 100-fold increase in throughput compared to the state of the art.

Clathrin coats partially preassemble and subsequently bend during endocytosis

Eukaryotic cells use clathrin-mediated endocytosis to take up a large range of extracellular cargo. During endocytosis, a clathrin coat forms on the plasma membrane, but it remains controversial when and how it is remodeled into a spherical vesicle. Here, we use 3D superresolution microscopy to determine the precise geometry of the clathrin coat at large numbers of endocytic sites. Through pseudo-temporal sorting, we determine the average trajectory of clathrin remodeling during endocytosis. We find that clathrin coats assemble first on flat membranes to 50% of the coat area before they become rapidly and continuously bent, and this mechanism is confirmed in three cell lines. We introduce the cooperative curvature model, which is based on positive feedback for curvature generation. It accurately describes the measured shapes and dynamics of the clathrin coat and could represent a general mechanism for clathrin coat remodeling on the plasma membrane.

Single molecule imaging simulations with advanced fluorophore photophysics

Advanced fluorescence imaging techniques such as single-molecule localization microscopy (SMLM) fundamentally rely on the photophysical behavior of the employed fluorophores. This behavior is generally complex and impacts data quality in a subtle manner. A simulation software named Single-Molecule Imaging Simulator (SMIS) is introduced that simulates a widefield microscope and incorporates fluorophores with their spectral and photophysical properties. With SMIS, data collection schemes combining 3D, multicolor, single-particle-tracking or quantitative SMLM can be implemented. The influence of advanced fluorophore characteristics, imaging conditions, and environmental parameters can be evaluated, facilitating the design of real experiments and their proper interpretation.

Benchmarking Thiolate-Driven Photoswitching of Cyanine Dyes

Carbocyanines are among the best performing dyes in single-molecule localization microscopy (SMLM), but their performance critically relies on optimized photoswitching buffers. Here, we study the versatile role of thiols in cyanine photoswitching at varying intensities generated in a single acquisition by a microelectromechanical systems (MEMS) mirror placed in the excitation path. The key metrics we have analyzed as a function of the thiolate concentration are photon budget, on-state and off-state lifetimes and the corresponding impact on image resolution. We show that thiolate acts as a concentration bandpass filter for the maximum achievable resolution and determine a minimum of ∼1 mM is necessary to facilitate SMLM measurements. We also identify a concentration bandwidth of 1-16 mM in which the photoswitching performance can be balanced between high molecular brightness and high off-time to on-time ratios. Furthermore, we monitor the performance of the popular oxygen scavenger system based on glucose and glucose oxidase over time and show simple measures to avoid acidification during prolonged measurements. Finally, the impact of buffer settings is quantitatively tested on the distribution of the glucose transporter protein 4 within the plasma membrane of adipocytes. Our work provides a general strategy for achieving optimal resolution in SMLM with relevance for the development of novel buffers and dyes.

Direct Stochastic Optical Reconstruction Microscopy (dSTORM) of Peroxisomes

Peroxisomes are central metabolic organelles whose maturation and function depend on efficient and accurate targeting of peroxisomal membrane proteins (PMPs). Ultrastructural imaging of the PMPs is a quite difficult task as it requires high spatial and temporal resolution. Further, the spatial resolution of conventional light microscopy is limited due to the diffraction of light. However, recent methodological developments in super resolution microscopy showed us to access the nanoscale regimes spatially allowing to elucidate the membrane structures of cell organelles. In this chapter, we present protocols used in our laboratory for the super-resolution imaging of the peroxisomal membrane protein 14 (PEX14p) by direct stochastic optical reconstruction microscopy (dSTORM).

Unveiling the Nanoscale Dynamics of the Exocytic Machinery in Chromaffin Cells with Single-Molecule Imaging

Neuronal and hormonal communication relies on the exocytic fusion of vesicles containing neurotransmitters and hormones with the plasma membrane. This process is tightly regulated by key protein-protein and protein-lipid interactions and culminates in the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex formation and zippering that promotes vesicular fusion. Located on both sides of the vesicle and the plasma membrane, the zippering of the SNARE complex acts to overcome the energy barrier afforded by the repulsive electrostatic force stemming from apposing two negatively charged phospholipid membranes. Another component opposing the timely organization of the fusion machinery is thermal Brownian energy that tends to homogenize all cellular molecules by constantly switching their motions and directions through short-lived molecular interactions. Much less is known of the mechanisms counteracting these chaotic forces, allowing seamless cellular functions such as exocytic fusion. Super-resolution microscopy techniques such as single-molecule imaging have proven useful to start uncovering these nanoscale mechanisms. Here, we used single-particle tracking photoactivatable localization microscopy (sptPALM) to track syntaxin-1-mEos, a SNARE protein located on the plasma membrane of cultured bovine chromaffin cells. We demonstrate that syntaxin-1-mEos undergoes dramatic change in its mobility in response to secretagogue stimulation leading to increased nanoclustering. These nanoclusters are transient in nature and likely to provide docked vesicles with a molecular environment conducive to exocytic fusion.

Maximum-likelihood model fitting for quantitative analysis of SMLM data

Quantitative data analysis is important for any single-molecule localization microscopy (SMLM) workflow to extract biological insights from the coordinates of the single fluorophores. However, current approaches are restricted to simple geometries or require identical structures. Here, we present LocMoFit (Localization Model Fit), an open-source framework to fit an arbitrary model to localization coordinates. It extracts meaningful parameters from individual structures and can select the most suitable model. In addition to analyzing complex, heterogeneous and dynamic structures for in situ structural biology, we demonstrate how LocMoFit can assemble multi-protein distribution maps of six nuclear pore components, calculate single-particle averages without any assumption about geometry or symmetry, and perform a time-resolved reconstruction of the highly dynamic endocytic process from static snapshots. We provide extensive simulation and visualization routines to validate the robustness of LocMoFit and tutorials to enable any user to increase the information content they can extract from their SMLM data.

ReCSAI: recursive compressed sensing artificial intelligence for confocal lifetime localization microscopy

BACKGROUND

Localization-based super-resolution microscopy resolves macromolecular structures down to a few nanometers by computationally reconstructing fluorescent emitter coordinates from diffraction-limited spots. The most commonly used algorithms are based on fitting parametric models of the point spread function (PSF) to a measured photon distribution. These algorithms make assumptions about the symmetry of the PSF and thus, do not work well with irregular, non-linear PSFs that occur for example in confocal lifetime imaging, where a laser is scanned across the sample. An alternative method for reconstructing sparse emitter sets from noisy, diffraction-limited images is compressed sensing, but due to its high computational cost it has not yet been widely adopted. Deep neural network fitters have recently emerged as a new competitive method for localization microscopy. They can learn to fit arbitrary PSFs, but require extensive simulated training data and do not generalize well. A method to efficiently fit the irregular PSFs from confocal lifetime localization microscopy combining the advantages of deep learning and compressed sensing would greatly improve the acquisition speed and throughput of this method.

RESULTS

Here we introduce ReCSAI, a compressed sensing neural network to reconstruct localizations for confocal dSTORM, together with a simulation tool to generate training data. We implemented and compared different artificial network architectures, aiming to combine the advantages of compressed sensing and deep learning. We found that a U-Net with a recursive structure inspired by iterative compressed sensing showed the best results on realistic simulated datasets with noise, as well as on real experimentally measured confocal lifetime scanning data. Adding a trainable wavelet denoising layer as prior step further improved the reconstruction quality.

CONCLUSIONS

Our deep learning approach can reach a similar reconstruction accuracy for confocal dSTORM as frame binning with traditional fitting without requiring the acquisition of multiple frames. In addition, our work offers generic insights on the reconstruction of sparse measurements from noisy experimental data by combining compressed sensing and deep learning. We provide the trained networks, the code for network training and inference as well as the simulation tool as python code and Jupyter notebooks for easy reproducibility.

Extending resolution within a single imaging frame

The resolution of fluorescence microscopy images is limited by the physical properties of light. In the last decade, numerous super-resolution microscopy (SRM) approaches have been proposed to deal with such hindrance. Here we present Mean-Shift Super Resolution (MSSR), a new SRM algorithm based on the Mean Shift theory, which extends spatial resolution of single fluorescence images beyond the diffraction limit of light. MSSR works on low and high fluorophore densities, is not limited by the architecture of the optical setup and is applicable to single images as well as temporal series. The theoretical limit of spatial resolution, based on optimized real-world imaging conditions and analysis of temporal image stacks, has been measured to be 40 nm. Furthermore, MSSR has denoising capabilities that outperform other SRM approaches. Along with its wide accessibility, MSSR is a powerful, flexible, and generic tool for multidimensional and live cell imaging applications.

Partition of estimated locations: an approach to accurate quality metrics for stochastic optical localization nanoscopy

Performance evaluation of localization algorithms in stochastic optical localization nanoscopy is necessary and important to applications. By simulation, a localization algorithm estimates a set of emitter locations from a simulated data movie, whose error in comparison with the set of true locations indicates the performance of the algorithm. Since the partition of estimated locations is unknown, the sample root mean square error (RMSE) cannot be computed, and the universal root mean square minimum distance (RMSMD) eventually becomes saturated as localization errors become large. In this paper, we propose a partition algorithm to estimate the partition of estimated locations. It makes use of three facts: (i) the true locations are known; (ii) the number of activations for each emitter is known; (iii) an estimated location is more likely to be associated with the nearest available emitter and vice versa. The estimated partition enables computation of the sample RMSE (RMSE-P) and improvement of the RMSMD with modification (RMSMD-P). Two simulations are carried out to demonstrate the efficacy of the partition algorithm and the metrics of RMSE-P and RMSMD-P. One investigates the effect of a large range of localization biases, and the other examines performance of the unbiased Gaussian information-achieving (UGIA) estimator. As shown by the results of both simulations, the proposed partition algorithm accurately estimates the partition in terms of the F1 score; with the partition estimated by the partition algorithm, the RMSE-P and RMSMD-P are approximately equal to the RMSE with the true partition in a large range of localization biases and errors. This demonstrates their broad applicability in performance evaluation of localization algorithms under the benchmark of the UGIA estimator.

Could artificial intelligence revolutionize the development of nanovectors for gene therapy and mRNA vaccines?

Gene therapy enables the introduction of nucleic acids like DNA and RNA into host cells, and is expected to revolutionize the treatment of a wide range of diseases. This growth has been further accelerated by the discovery of CRISPR/Cas technology, which allows accurate genomic editing in a broad range of cells and organisms in vitro and in vivo. Despite many advances in gene delivery and the development of various viral and non-viral gene delivery vectors, the lack of highly efficient non-viral systems with low cellular toxicity remains a challenge. The application of cutting-edge technologies such as artificial intelligence (AI) has great potential to find new paradigms to solve this issue. Herein, we review AI and its major subfields including machine learning (ML), neural networks (NNs), expert systems, deep learning (DL), computer vision and robotics. We discuss the potential of AI-based models and algorithms in the design of targeted gene delivery vehicles capable of crossing extracellular and intracellular barriers by viral mimicry strategies. We finally discuss the role of AI in improving the function of CRISPR/Cas systems, developing novel nanobots, and mRNA vaccine carriers.

Electron-beam patterned calibration structures for structured illumination microscopy

Super-resolution fluorescence microscopy can be achieved by image reconstruction after spatially patterned illumination or sequential photo-switching and read-out. Reconstruction algorithms and microscope performance are typically tested using simulated image data, due to a lack of strategies to pattern complex fluorescent patterns with nanoscale dimension control. Here, we report direct electron-beam patterning of fluorescence nanopatterns as calibration standards for super-resolution fluorescence. Patterned regions are identified with both electron microscopy and fluorescence labelling of choice, allowing precise correlation of predefined pattern dimensions, a posteriori obtained electron images, and reconstructed super-resolution images.

Particle Localization Using Local Gradients and Its Application to Nanometer Stabilization of a Microscope

Particle localization plays a fundamental role in advanced biological techniques such as single-molecule tracking, superresolution microscopy, and manipulation by optical and magnetic tweezers. Such techniques require fast and accurate particle localization algorithms as well as nanometer-scale stability of the microscope. Here, we present a universal method for three-dimensional localization of single labeled and unlabeled particles based on local gradient calculation of particle images. The method outperforms state-of-the-art localization techniques in high-noise conditions, and it is capable of 3D nanometer accuracy localization of nano- and microparticles with sub-millisecond calculation time. By localizing a fixed particle as fiducial mark and running a feedback loop, we demonstrate its applicability for active drift correction in sensitive nanomechanical measurements such as optical trapping and superresolution imaging. A multiplatform open software package comprising a set of tools for local gradient calculation in brightfield, darkfield, and fluorescence microscopy is shared for ready use by the scientific community.

Enabling Spectrally Resolved Single-Molecule Localization Microscopy at High Emitter Densities

Single-molecule localization microscopy (SMLM) is a powerful super-resolution technique for elucidating structure and dynamics in the life- and material sciences. Simultaneously acquiring spectral information (spectrally resolved SMLM, sSMLM) has been hampered by several challenges: an increased complexity of the optical detection pathway, lower accessible emitter densities, and compromised spatio-spectral resolution. Here we present a single-component, low-cost implementation of sSMLM that addresses these challenges. Using a low-dispersion transmission grating positioned close to the image plane, the +1^stdiffraction order is minimally elongated and is analyzed using existing single-molecule localization algorithms. The distance between the 0^th and 1^st order provides accurate information on the spectral properties of individual emitters. This method enables a 5-fold higher emitter density while discriminating between fluorophores whose peak emissions are less than 15 nm apart. Our approach can find widespread use in single-molecule applications that rely on distinguishing spectrally different fluorophores under low photon conditions.

splitSMLM, a spectral demixing method for high-precision multi-color localization microscopy applied to nuclear pore complexes

Single molecule localization microscopy (SMLM) with a dichroic image splitter can provide invaluable multi-color information regarding colocalization of individual molecules, but it often suffers from technical limitations. Classical demixing algorithms tend to give suboptimal results in terms of localization precision and correction of chromatic errors. Here we present an image splitter based multi-color SMLM method (splitSMLM) that offers much improved localization precision and drift correction, compensation of chromatic distortions, and optimized performance of fluorophores in a specific buffer to equalize their reactivation rates for simultaneous imaging. A novel spectral demixing algorithm, SplitViSu, fully preserves localization precision with essentially no data loss and corrects chromatic errors at the nanometer scale. Multi-color performance is further improved by using optimized fluorophore and filter combinations. Applied to three-color imaging of the nuclear pore complex (NPC), this method provides a refined positioning of the individual NPC proteins and reveals that Pom121 clusters act as NPC deposition loci, hence illustrating strength and general applicability of the method.

The development of an image splitter based multi-colour single-molecule localization microscopy method (splitSMLM) in combination with a spectral demixing algorithm improves localization accuracy as exemplified by three-colour imaging of nuclear pore complex proteins.

GDSC SMLM: Single-molecule localisation microscopy software for ImageJ

Single-molecule localisation microscopy (SMLM) uses software to extract super-resolved positions from microscope images of fluorescent molecules. These localisations can then be used to render super-resolution images or analysed to extract information about molecular behaviour. The GDSC SMLM software provides a set of tools for analysing SMLM data in a single cross-platform environment. The software identifies fluorescent molecules in raw microscope images and localises their positions using stages of spot detection, spot fitting and spot rejection. The resulting localisation data set can then be visualised, cropped and filtered. A suite of downstream analysis tools enable the user to perform single-particle tracking, cluster analysis and drift correction. In addition, GDSC SMLM also provides utility tools that enable modelling of EM-CCD and sCMOS cameras as well as point spread functions (PSFs) for data simulation. The software is written in Java and runs as a collection of plugins for the ImageJ software.

Towards optimal point spread function design for resolving closely spaced emitters in three dimensions

The past decade has brought many innovations in optical design for 3D super-resolution imaging of point-like emitters, but these methods often focus on single-emitter localization precision as a performance metric. Here, we propose a simple heuristic for designing a point spread function (PSF) that allows for precise measurement of the distance between two emitters. We discover that there are two types of PSFs that achieve high performance for resolving emitters in 3D, as quantified by the Cramér-Rao bounds for estimating the separation between two closely spaced emitters. One PSF is very similar to the existing Tetrapod PSFs; the other is a rotating single-spot PSF, which we call the crescent PSF. The latter exhibits excellent performance for localizing single emitters throughout a 1-µm focal volume (localization precisions of 7.3 nm in x, 7.7 nm in y, and 18.3 nm in z using 1000 detected photons), and it distinguishes between one and two closely spaced emitters with superior accuracy (25-53% lower error rates than the best-performing Tetrapod PSF, averaged throughout a 1-µm focal volume). Our study provides additional insights into optimal strategies for encoding 3D spatial information into optical PSFs.

Deep-SMOLM: deep learning resolves the 3D orientations and 2D positions of overlapping single molecules with optimal nanoscale resolution

Dipole-spread function (DSF) engineering reshapes the images of a microscope to maximize the sensitivity of measuring the 3D orientations of dipole-like emitters. However, severe Poisson shot noise, overlapping images, and simultaneously fitting high-dimensional information-both orientation and position-greatly complicates image analysis in single-molecule orientation-localization microscopy (SMOLM). Here, we report a deep-learning based estimator, termed Deep-SMOLM, that achieves superior 3D orientation and 2D position measurement precision within 3% of the theoretical limit (3.8° orientation, 0.32 sr wobble angle, and 8.5 nm lateral position using 1000 detected photons). Deep-SMOLM also demonstrates state-of-art estimation performance on overlapping images of emitters, e.g., a 0.95 Jaccard index for emitters separated by 139 nm, corresponding to a 43% image overlap. Deep-SMOLM accurately and precisely reconstructs 5D information of both simulated biological fibers and experimental amyloid fibrils from images containing highly overlapped DSFs at a speed ~10 times faster than iterative estimators.