Super-Resolution Radial Fluctuations (SRRF): A Versatile and Accessible Tool for Live-Cell Nanoscopy and Multimodal Imaging

Super-Resolution Radial Fluctuations (SRRF): A Versatile and Accessible Tool for Live-Cell Nanoscopy and Multimodal Imaging | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 17 March 2025 V1 Latest version Share on Super-Resolution Radial Fluctuations (SRRF): A Versatile and Accessible Tool for Live-Cell Nanoscopy and Multimodal Imaging Authors : Sanhua Fang 0000-0003-0525-642X [email protected] , Li Liu , Dan Yang , Shuangshuang Liu , and Qiong Huang Authors Info & Affiliations https://doi.org/10.22541/au.174224393.37284045/v1 401 views 222 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract To break through the diffraction limit of light, various super-resolution techniques have been developed. Super-resolution Radial Fluctuations (SRRF) is an emerging super-resolution microscopy technique that utilizes a standard wide-field fluorescence microscope and open-source software plugins compatible with ImageJ. As a result, SRRF has relatively low hardware and software requirements, making it highly accessible to researchers. Here we first introduce the basic principles and workflow of SRRF, then describe the open-source ImageJ software tools NanoJ-SRRF and NanoJ-SQUIRREL. NanoJ-SRRF is used for reconstructing super-resolution images, while NanoJ-SQUIRREL is used for quantitatively assessing the quality of super-resolution images. Next, We summarize the advantages and disadvantages of SRRF as well as optimization methods. Finally, we present the applications of SRRF technology, along with potential avenues for future technical improvements. Super-Resolution Radial Fluctuations (SRRF): A Versatile and Accessible Tool for Live-Cell Nanoscopy and Multimodal Imaging Sanhua Fang, Li Liu, Dan Yang, Shuangshuang Liu, Qiong Huang （Core Facilities, Zhejiang University School of Medicine, Hangzhou 310058, China） Abstract: To break through the diffraction limit of light, various super-resolution techniques have been developed. Super-resolution Radial Fluctuations (SRRF) is an emerging super-resolution microscopy technique that utilizes a standard wide-field fluorescence microscope and open-source software plugins compatible with ImageJ. As a result, SRRF has relatively low hardware and software requirements, making it highly accessible to researchers. Here we first introduce the basic principles and workflow of SRRF, then describe the open-source ImageJ software tools NanoJ-SRRF and NanoJ-SQUIRREL. NanoJ-SRRF is used for reconstructing super-resolution images, while NanoJ-SQUIRREL is used for quantitatively assessing the quality of super-resolution images. Next, We summarize the advantages and disadvantages of SRRF as well as optimization methods. Finally, we present the applications of SRRF technology, along with potential avenues for future technical improvements. Key words: Super-resolution Radial Fluctuations (SRRF); Principle; Advances; Applications Correspondence Sanhua Fang, Core Facilities, Zhejiang University School of Medicine, Hangzhou 310058, China. Email: [email protected] Acknowledgements: the work is supported by Zhejiang Province Basic Public Welfare Program (LGC19H090001), Scientific Research Project of Zhejiang Provincial Department of Education (Y202250110) and Zhejiang University Instrument Development and Cultivation Program( YQZX-C202412). 1 | INTRODUCTION Fluorescence microscopy enables the dynamic observation of the behavior of specifically labeled molecules and organelles within live cells, greatly promoting the development of cell biology. However, the spatial resolution of traditional fluorescence microscopy is typically limited by the optical diffraction limit, with lateral resolution around 200-300 nm and axial resolution around 500-700 nm. Therefore, the study of smaller cellular components has relied on electron microscopy, which involves complex sample preparation and is generally used for fixed samples, making it unsuitable for live biological specimens. In the early 21st century, a new family of microscopy known as super-resolution microscopy emerged. Super-resolution microscopy overcome the optical diffraction limit, achieving resolutions close to tens of nanometers. This advanced microscopy has significant advantages, offering resolutions comparable to electron microscopy while retaining the benefits of optical microscopy, such as the availability of a wide range of highly specific molecular tags, simple sample preparation, and compatibility with live cell imaging. Due to their groundbreaking contributions, the developers of super-resolution microscopy were awarded the Nobel Prize in Chemistry in 2014. 1 Super-resolution microscopy include Stimulated Emission Depletion (STED) microscopy, 2 Single Molecule Localization Microscopy (SMLM), 3 and Structured Illumination Microscopy (SIM). 4 STED microscopy is based on confocal scanning technology, introducing a second doughnut-shaped laser beam into the illumination path to suppress fluorescence from the outer regions of the excitation point spread function (PSF), thereby reducing the size of the PSF and improving resolution. The resolution of STED microscopy is proportional to the intensity of the doughnut beam, achieving resolutions of less than 100 nm. However, it requires beam intensities of 0.1-1 GW/cm², resulting in significant phototoxicity, which limits its application in live cell imaging. SMLM can achieve spatial resolutions of 30 nm. However, it typically requires thousands to tens of thousands of frames for high-density reconstruction, improving spatial resolution at the expense of temporal resolution. The reconstruction algorithm for SMLM relies on the mathematical calculation of isolated single-molecule emissions, necessitating laser intensities of KW/cm²to reduce the overlap of emitting molecules. Prolonged exposure also results in significant phototoxicity, generally used for fixed samples. On the other hand, SIM has low illumination requirements, fast imaging speed, and compatibility with traditional fluorophores, making it an attractive method for live-cell super-resolution imaging. However, the resolution of SIM is lower than that of STED and SMLM, and it is challenging for SIM to achieve resolutions less than 100 nm. Super-resolution microscopy have significantly promoted the study of subcellular fine structures and dynamic changes. However, STED and SMLM typically require high laser intensities, leading to high phototoxicity and limited applicability for live-cell samples. While SIM can be used for super-resolution imaging of live cells, it requires expensive optical hardware. Therefore, there is a substantial practical demand for achieving super-resolution imaging using conventional wide-field fluorescence microscopes. Recently, computational imaging methods have gained increasing attention, such as deconSTORM, 5 3B, 6 SOFI, 7 and SRRF. 8 The name deconSTORM comes from ”deconvolution of STORM data,” which involves deconvolution processing of STORM data, optimizing the super-resolution imaging. The name 3B stands for ”Bayesian analysis of Blinking and Bleaching,” a technique that uses Bayesian statistical methods to analyze the blinking and bleaching phenomena of fluorescent molecules during experiments. Both deconSTORM and 3B are commonly used to enhance the localization accuracy and resolution of SMLM. Super-resolution optical fluctuation imaging (SOFI) and super-resolution radial fluctuations (SRRF) utilize the fluctuation information of fluorescence signals to overcome the traditional resolution limits of optical microscopes. Both methods do not rely on complex experimental setups or special labeling and can be applied on existing conventional fluorescence microscopy platforms. By analyzing the temporal fluctuation information of fluorescence, they enhance imaging resolution, making them suitable for studying dynamic biological processes with the advantages of lower phototoxicity and reduced photobleaching. Compared to SOFI, SRRF offers faster imaging speeds, simpler and quicker computations, higher flexibility and convenience, as well as strong adaptability to low signal-to-noise ratios images. Therefore, SRRF is an efficient, real-time, and highly adaptable super-resolution imaging technique particularly suited for dynamic study of biological processes. SRRF is expected to greatly advance research on subcellular ultrastructures and dynamic changes. Therefore, this review will introduce the principles, procedures, implementation methods, advantages and disadvantages, applications, and future development directions of SRRF, promoting its application in cell biology. 2 | PRICINPLE SRRF is a computational imaging technique that analyzes radial fluctuations using time-series images. It does not require the detection and localization of individual fluorescent molecules. Instead, it calculates the local gradient convergence (referred to as radiality) of the entire image frame. 8 SRRF includes Spatial and temporal analysis (Figure 1) . Fluorescently labeled molecules exhibit random fluctuations at different spatial positions. These fluctuations typically display radial symmetry in space, meaning that the signal intensity fluctuations around the molecule’s center follow a certain radial pattern. SRRF magnifies each pixel into ”sub-pixels,” with each sub-pixel assigned a non-binary value related to the probability of containing a fluorophore. To calculate this value, SRRF measures the local radial symmetry within the image. This radial distribution can distinguish between two points with Gaussian point spread functions (PSFs) that have full-width at half-maximum (FWHM) of 0.7 times the original. The FWHM of the radial distribution can be adjusted by changing the radius over which the gradient convergence is measured. This distribution is independent of the PSF intensity and robust to changes in PSF size. Thus, this method enhances image resolution by analyzing radial fluctuations. Additionally, the radiality of a complete image will include some radial peaks unrelated to fluorophores, as transient local radial symmetry may appear within the image noise. By conducting time-series analysis, these transient local radial symmetries can be removed, thereby improving the image’s contrast and resolution. 3 | Workflow Based on its principle, the SRRF algorithm consists of two parts: spatial analysis and temporal analysis. Spatial analysis uses the fluorescence distribution of each frame to generate a series of radiality maps, while temporal analysis combines these radiality maps into a single SRRF super-resolution image through temporal statistics. Therefore, the SRRF workflow (Figure 1) includes image acquisition, spatial analysis, temporal analysis, and super-resolution image reconstruction. 9 3.1 | Image acquisition The first step of SRRF involves acquiring time series images using a conventional wide-field fluorescence microscope, total internal reflection fluorescence microscope, or laser confocal microscope. These images are usually captured at short intervals, with each frame capturing the fluorescent molecules within cells or samples.Key parameters like frame rate, exposure time, and illumination intensity are optimized to balance temporal resolution, signal-to-noise ratio (SNR), and photobleaching. This raw dataset provides the spatiotemporal information needed for subsequent super-resolution processing. 3.2 | Spatial analysis Spatial analysis identifies sub-diffraction features by evaluating radial intensity gradients in each frame. SRRF magnifies each pixel into subpixels, and each is given a value related to the probability of it containing a fluorophore. This probability is based on the radial symmetry or ’radiality’ that fluorophore PSFs have, due to being circular in nature. Radiality is measured for each subpixel in the image by taking a ring of neighboring subpixels and measuring the convergence of intensity gradients passing through these. Subpixels closer to the true center of molecules have higher radiality values, whereas subpixels far from molecule centers have lower radiality values. 3.3 Temporal analysis The radiality of a complete image will include some radial peaks unrelated to fluorophores, as transient local radial symmetry may appear within the image noise. By tracking the radiality over time, it is clear if a subpixel contains temporally correlated signal, or temporally uncorrelated signal (noise). So these transient local radial symmetries can be removed, thereby improving the image’s contrast and resolution. 3.4 Super-Resolution Image Reconstruction Using iterative algorithms, SRRF reconstructs a high-resolution image by combining radiality metrics, temporal correlation maps, and raw intensity data. The output image surpasses the diffraction limit, typically achieving 2–5×resolution improvement (e.g., compatibility due to lower light exposure and faster processing than SMLM. The result is a detailed, dynamic visualization of subcellular structures, such as cytoskeletal networks or organelle interactions, with minimal photodamage. 4 | NanoJ-SRRF The SRRF algorithm is provided in the form of the NanoJ-SRRF software package, a free and widely used open-source plugin for ImageJ or FIJI image analysis software. Parameter settings in NanoJ-SRRF significantly impact the effectiveness of SRRF super-resolution reconstruction. 8,9 4.1 | Ring radius The ring radius defines the size of the sampling area around the central pixel when calculating local fluctuations. Adjusting the ring radius can influence the SRRF algorithm’s response to local structures in the image and the extraction of super-resolution information. Generally, an appropriate ring radius is selected based on the image resolution, noise level, and specific characteristics of the sample. With higher image resolution, a smaller ring radius can better capture subtle fluctuations; in noisier images, increasing the ring radius can help smooth out noise and avoid over-response to noise fluctuations. If the sample structures are large, appropriately increasing the ring radius can provide more local information, enhancing the super-resolution effect. 4.2 | Radiality magnification Radiality magnification controls the scale of radial fluctuations in the image. Increasing this value magnifies local details in the image, making smaller structures clearer in the reconstructed image. However, excessive magnification may also amplify noise. For more complex samples (e.g., smaller fluorescent particles or intracellular microstructures), appropriate magnification can significantly improve detail visibility, but care must be taken to avoid over-magnification, which can cause image blurring or artifacts. Higher radiality magnification increases computational complexity, so a balance must be keep between computational resources and the required precision. 4.3 | Axes in ring The number of axes in the ring refers to the number of components within the ring region analyzed for radial fluctuations in SRRF. This parameter controls the number of spatial features in the local region, thereby affecting the accuracy and detail of the reconstruction. For complex image structures or when finer resolution is needed, increasing the number of ring axes usually helps better reconstruct details. However, in noisier images, a high number of ring axes may amplify noise. Additionally, a higher number of ring axes increases computational complexity, particularly in large or high-resolution datasets. 4.4 | Temporal radiality analysis Temporal radiality parameters include Temporal Radiality Maximum (TRM), Temporal Radiality Average (TRA), Temporal Radiality Pairwise Product Mean (TRPPM), and Temporal Radiality Auto-Correlations (TRAC). These parameters are used to analyze and optimize SRRF results based on time-series data, helping to extract super-resolution information from dynamic image data. TRM is primarily used to capture local fluctuations’ extremities in the time series, helping to identify the most significant changes or peaks. It is typically used for sparse datasets (e.g., classic PALM/STORM experiments). TRA analyzes the average level of radial fluctuations over a period, helping to determine the average fluctuation intensity of the sample at each time point. TRA is particularly suitable for blinking datasets, where fluorescent molecules may overlap or originate from total internal reflection fluorescence or wide-field fluorescence microscopy data with limited intensity fluctuations. TRPPM evaluates the correlation of radial fluctuations between multiple time points by calculating the product of fluctuations at different time points. It measures the similarity or dependence between two time points, especially useful for tracking local features in the time series. The images generated by TRPPM contain details and resolution similar to TRA but can provide additional noise suppression. TRAC is an autocorrelation function used to calculate the correlation between radial fluctuations at each time point and its preceding time points, assessing the autocorrelation of local features in the time dimension. TRAC performs autocorrelation analysis similar to SOFI super-resolution techniques and performs well in dense datasets with overlapping fluorescent molecules. 4.5 | Radiality Radiality parameters include Remove Positivity Constraint, Renormalize, and Do Gradient Smoothing. Whether to enable these parameters depends on data type and image quality. Enabling Remove Positivity Constraint allows SRRF reconstruction to no longer restrict the output image’s pixel intensity to positive values. Removing the positivity constraint can allow the SRRF reconstruction algorithm to fit the data more freely, potentially recovering more image details, especially in cases of strong background noise or weak signals. However, if image quality is high, artifacts or unnatural negative values may appear. The Renormalize option involves standardizing SRRF reconstruction results to fit a certain intensity or brightness range, ensuring the dynamic range of the output image is appropriate, preventing overexposure or underexposure. Without choosing the Renormalize option, SRRF tends to enhance edges in the image. The Do Gradient Smoothing option applies gradient smoothing during SRRF reconstruction, helping to remove noise and smooth image edges, especially useful for noisier images. If the image quality is high, noise is low, or you wish to preserve image details and edge information, disabling this option can retain more details and avoid over-smoothing. 4.6 | Weighting Weighting parameters include Do Intensity Weighting and Do Gradient Weighting. These parameters adjust how intensity and gradient information in the image is handled during super-resolution reconstruction. Intensity Weighting weights each pixel based on the fluorescence molecule’s intensity during SRRF reconstruction, meaning higher intensity pixels will have more importance in the reconstruction. The spatial part of SRRF analysis—the radiality map—contains many small radial peaks within the noise. When performing subsequent temporal analysis (e.g., TRPPM or TRAC), these peaks will be suppressed. However, using TRM or TRA to display the image will not suppress these peaks. Therefore, intensity weighting is applied to denoise TRM/TRA images, enhancing radial peaks from high-intensity regions. For TRPPM/TRAC analysis, the effect of intensity weighting is less critical. Gradient Weighting weights different regions based on the pixel gradient. Areas with significant gradient changes (e.g., edges or structures in the image) receive more weight, appearing clearer in the reconstructed image. Gradient weighting increases the contrast of point-like objects to reduce noise, making it particularly useful in datasets where blinking is observed. Its effect is less pronounced in relatively dense data collected from conventional wide-field fluorescence microscopy. 5 | NanoJ-SQUIRREL Super-resolution techniques rely on multiple steps that may lead to the formation of image artifacts, resulting in the misinterpretation of biological information. SQUIRREL (Super-Resolution Quantitative Image Rating and Reporting of Error Locations) is an ImageJ-based analysis method that quantitatively evaluates the quality of super-resolution images. By comparing diffraction-limited images with super-resolution images of the same acquisition volume, it generates quantitative maps of super-resolution errors and helps researchers optimize imaging parameters. 10 NanoJ-SQUIRREL is part of the NanoJ series and aims to help users better understand potential issues in image data and improve the efficiency and accuracy of SRRF by precisely locating and quantitatively analyzing error locations in super-resolution images. NanoJ-SQUIRREL has two core functions: error location reporting and quantitative image scoring. In SRRF imaging, various factors such as fluorescent labeling, detector errors, and image reconstruction algorithms can cause localization errors, which affect the localization accuracy in the image and thus the reliability of experimental results. The error location reporting function of NanoJ-SQUIRREL can accurately identify and mark the locations of errors in the image, providing researchers with detailed error analysis reports. In terms of quantitative image scoring, NanoJ-SQUIRREL calculates two global image quality metrics: RSE (Resolution Scale Error), which is the root mean square error between the reference image and the resolution scale image; and RSP (Resolution Scale Pearson coefficient), which is the Pearson correlation coefficient between the reference image and the resolution scale image. SQUIRREL is sensitive not only to the disappearance of structures but also to common super-resolution artifacts, including merged structures and bright aggregates. 6 | ADVANTAGES AND DISADVANTAGES SRRF is based on standard wide-field fluorescence microscopy and uses small radial fluctuation signals to improve the spatial resolution of images, generating super-resolution images. Compared to super-resolution microscopy such as STED, STORM, and SIM (table 1), SRRF possesses numerous advantages but also has some limitations . 9,11 6.1 | Advantages 6.1.1 | Less requirement for expensive equipment SRRF does not require costly specialized equipment. It can achieve super-resolution image reconstruction using a conventional wide-field fluorescence microscope and open-source software plugins, making it highly accessible to life science researchers. 6.1.2 No need for switchable fluorescent molecules Unlike single-molecule localization techniques like STORM and PALM, SRRF does not rely on the switching behavior of fluorescent molecules (dynamic changes between the on and off states). SRRF can use conventional fluorescent dyes or markers, greatly simplifying sample preparation and experimental procedures. This is particularly beneficial in cases where special switchable molecules cannot be used, providing a broader range of applications. 6.1.3 | Low phototoxicity and low photobleaching SRRF does not depend on intense switching of fluorescent molecules and sparse distribution of emitting fluorophores. Instead, it enhances image resolution by analyzing the natural fluctuations of fluorescent molecules. This allows SRRF to use lower excitation light intensity during imaging, reducing photodamage and photobleaching of the sample. This is especially important for long-term imaging or dynamic observation of biological samples. 6.1.4 | Dynamic imaging and multi-modal Imaging Methods like STORM and PALM typically require thousands to tens of thousands of image frames to reconstruct images by sparse localization of markers, whereas the SRRF algorithm processes short image sequences to generate super-resolution images in real-time. SRRF has lower computational complexity, and can significantly enhance imaging speed, making it particularly suitable for real-time observation of dynamic processes. SRRF can achieve super-resolution images based on wide-field fluorescence microscopy and can be easily integrated with electron microscopy, 12 expansion microscopy, 13 electrochemical luminescence microscopy, 14 and ultrasound technology, 15 enabling multi-modal imaging. 6.1.5 | Good compatibility and flexibility SRRF is highly compatible with existing microscopy hardware. Users can apply the SRRF algorithm on conventional confocal microscopes, wide-field microscopes, total internal reflection fluorescence microscopes, light-sheet microscopes, 16 and other devices, flexibly handling different types of image data. Additionally, SRRF can be combined with other super-resolution imaging techniques, such as structured illumination microscopy (SIM) or optical frequency modulation microscopy (FPM), to achieve higher resolution and richer information acquisition. 17 6.2 | Disadvantages 6.2.1 | Resolution Improvement Limitations Although SRRF can surpass the resolution limits of optical microscopes, its resolution enhancement is not unlimited. SRRF typically provides a resolution of about 50-100 nm, which still lags behind other super-resolution techniques like STORM and PALM. STORM and PALM can theoretically achieve higher resolutions (down to 20 nm) through single-molecule localization imaging. Therefore, SRRF may not meet the demands requiring extreme resolution. Additionally, the current iterations of the SRRF algorithm do not improve axial (z-axis) resolution, although it is compatible with techniques that enhance axial resolution, such as point-scanning confocal, total internal reflection fluorescence (TIRF), and spinning disk confocal microscopy. 6.2.2 | Artifacts and loss of detail SRRF may produce artifacts and loss of detail when studying high-density subcellular structures. The SRRF algorithm tends to inaccurately connect discretely distributed structures. When the density exceeds 30/μm², traditional SRRF algorithms cannot accurately reconstruct detailed fluorescence distributions, leading to inaccurate reconstructions. This may be due to the influence of radiality on adjacent fluorescent emitters when individual fluorophores undergo temporal intensity fluctuations. 6.2.3 | Imaging speed limitations SRRF requires at least hundreds of image frames to reconstruct super-resolution images, making it less suitable for live-cell imaging of rapid subcellular or molecular motion. 7 | OPTIMIZATION OF SRRF IMAGING Given the current limitations of SRRF, such as artifacts and loss of detail, resolution and imaging speed limitations, many researchers have further optimized SRRF technology to improve image fidelity, resolution, and imaging speed. These optimizations include gmSRRF, 18 eSRRF, 19 VeSRRF, 20 among others. 7.1 | gmSRRF The Gradient Magnitude Variance Corrected SRRF (gmSRRF) algorithm analyzes the temporal variation properties of intensity and gradient variance in the original image sequence. By analyzing high-density stochastic optical reconstruction microscopy (STORM) data as well as traditional wide-field, confocal, or SIM imaging sequences, gmSRRF reduce the resolution loss in SRRF images caused by artifacts, significantly enhancing the resolution and quality of the reconstructed images. 7.2 | Enhanced SRRF (eSRRF) Enhanced SRRF (eSRRF) is an improved version of the original SRRF algorithm that significantly enhances image fidelity and resolution. It makes three key modifications to the original algorithm to improve image fidelity: Sub-pixel generation using Fourier transform interpolation to minimize macro-pixel artifacts,Radial Gradient Convergence (RGC) for convergence averaging calculations, mapped based on a user-defined radius R weight factor. Integration of the SQUIRREL tool for artifact detection and quantification, providing automatic data-driven optimal parameter identification, minimizing image artifacts and nonlinearity. Additionally, eSRRF, when combined with confocal microscopy, extends its application to three-dimensional space, achieving live-cell 3D super-resolution imaging with a capture speed of one 3D image per second. 7.3 | VeSRRF VeSRRF (Variance Reweighted Radial Fluctuations and Enhanced SRRF) is a computational super-resolution imaging method that integrates intensity and gradient variance-weighted radial fluctuations (VRRF) with the eSRRF algorithm. This combination significantly improves the accuracy and robustness of super-resolution imaging, especially under low signal-to-noise ratio and high-noise conditions. VRRF first uses statistical analysis of intensity and gradient variance to separate overlapping fluorescent molecules and reduce artifacts caused by high-density fluorescent molecules. The processed image sequence is then further analyzed using the eSRRF algorithm, which enhances image fidelity through Radial Gradient Convergence (RGC) transformation. Compared to other algorithms in single-molecule localization microscopy (SMLM) and FF-SRM, VeSRRF consistently achieves the highest resolution and exceptional fidelity, outperforming SRRF, VRRF/SRRF, and eSRRF algorithms. 7.4 | SRRF imaging speed improvement To achieve real-time imaging and super-resolution image reconstruction, Ricardo Henriques developed SRRF-Stream (Super-Resolution Radial Fluctuations Stream), a dynamic extension of the SRRF algorithm. SRRF-Stream continuously assesses super-resolution image quality during acquisition and ensures optimal image quality by automatically adjusting acquisition parameters. This reduces acquisition time and data storage requirements. SRRF-Stream is uniquely advantageous in dynamic biological research, allowing real-time observation of intracellular molecular and subcellular structure dynamics without waiting for complex image reconstruction processes. 21 Yubing Han and colleagues used Graphics Processing Units (GPU) to accelerate the algorithm process and programmed in Python, enhancing the universality and computational speed of the SRRF algorithm, further applying it to various live-cell super-resolution microscopy methods. 17 Xu YK and colleagues applied a U-Net neural network architecture to accelerate SRRF reconstruction, achieving super-resolution SRRF reconstruction comparable to traditional methods using only five low SNR frames. 22 This significantly reduces the number of raw images required for super-resolution reconstruction, making real-time, long-term, multi-color live-cell super-resolution imaging possible. The authors achieved super-resolution visualization of dynamic microtubule shrinkage and interactions between microtubules and clathrin-coated pits (CCPs) using deep learning-accelerated SRRF. Multi-color live-cell super-resolution SRRF imaging can be easily achieved with low-dose illumination and conventional wide-field microscopy settings. 8 | APPLICATION 8.1 | High-resolution imaging of cellular ultrastructure SRRF breaks through the diffraction limit of light, achieving a resolution of approximately 100 nm, which allows scientists to observe in detail the organelles and their interactions, such as microtubules, actin filaments, mitochondria, lysosomes, and the Golgi apparatus. We captured time-lapse images of tubulin, Golgi apparatus, mitochondria, and nuclear pore complex using a spinning disk confocal microscope. These images were then processed with SRRF reconstruction to generate super-resolution images, achieving a 4.3- to 7-fold improvement in image resolution (Figure 2). Anjum et al. utilized SRRF to reveal significant changes in the heterochromatin structure of HEK 293A cells treated with ethyl methanesulfonate (EMS), as well as morphological changes in lysosomes and mitochondria. EMS treatment caused the heterochromatin structure in HEK 293A cells to become more dispersed and sparse, while lysosomes exhibited an elongated shape. Mitochondria also underwent significant structural changes, from a normal tubular structure to a more loose and fragmented form, and their area within the cytoplasm significantly increased. 23 Gilleron J et al. studied the localization and interaction of molecules such as Rab4b, VPS52, and CI-M6PR in cells using fluorescence labeling and SRRF. Rab4b regulated the formation of VPS52 microdomains, which imparted directional specificity to the retrograde transport of CI-M6PR. This mechanism relies on the GARP complex, ensuring that endosome-derived transport carriers correctly associate with and fuse with the Golgi apparatus. 24 SRRF is also used in plant cell wall ultrastructure studies. Donaldson LA et al. employed SRRF combined with confocal fluorescence microscopy to achieve super-resolution imaging of the xylem cell walls in Douglas-fir trees. By comparing spontaneous fluorescence imaging of lignin with staining using alanine or rhodamine B, SRRF achieved moderate resolution improvement in both spontaneous fluorescence and alanine staining, while achieving significant resolution enhancement in rhodamine B staining, reaching a sub-100-nanometer resolution. 25 8.2 Dynamic imaging of cellular ultrastructure The high-resolution imaging of live cells is very important for biomedical research. However, traditional super-resolution imaging techniques such as PALM and STORM, while offering high resolution, induce significant phototoxicity, limiting their application in long-term live-cell imaging. SRRF can achieve a resolution of over 150 nm under low light intensity and is compatible with standard wide-field, confocal, or TIRF microscopy, significantly reducing phototoxicity, making it suitable for high-resolution live-cell imaging. Nils Gustafsson et al. used SRRF to perform super-resolution imaging of HeLa cells expressing tubulin-GFP, achieving a resolution of 60 nm with a frame rate of one per second . They also studied the rapid remodeling of the actin cortex during immune synapse formation (Figure 2). 8 Yuan-Hao Lee et al. applied the SRRF algorithm to perform spatial correlation analysis of calcium signals in the mouse retina, and then studied the temporal changes in calcium signals through HeLa cells expressing Cx36-GCaMP calcium indicators. Using ImageJ and Matlab for image processing and analysis, SRRF reconstruction improved image resolution, enabling precise identification of calcium signals. 26 SRRF is highly compatible with fluorescent dyes. Muthukumaran et al. combined fluorescent-activated and absorbed transfer (FAST) tags with organic fluorescent markers and used SRRF to successfully observe the dynamic changes of mitochondria and microtubules in cells, achieving multi-color dynamic imaging. By supplementing fluorescent ligands, FAST can restore fluorescence intensity and resolution after photobleaching, making it an ideal tool for long-term continuous super-resolution imaging based on SRRF. 27 Carbon nanodots (CNDs), known for their excellent biocompatibility and fluorescence properties, have recently become an ideal choice for biological labeling. In particular, red-emitting CNDs have attracted widespread attention for their self-labeling capability without the need for further modification. Garg R et al. developed a novel red-emitting, highly photostable, water-soluble carbon nanodot (TPP CND) for super-resolution radial fluctuation (SRRF) microscopy to observe the dynamic changes of mitochondria. Using SRRF, the researchers successfully captured super-resolution structural changes of mitochondria under normal and hyperglycemic conditions, revealing significant mitochondrial fission under hyperglycemia, leading to the formation of many small and fragmented mitochondria. Additionally, they observed that after treatment with the anti-diabetic drug metformin, mitochondrial fission was enhanced, and under conditions of reduced mitochondrial membrane potential, TPP CNDs migrated from mitochondria to the cell nucleus. 30 8.3 | In Vivo imaging SRRF is highly compatible with fluorescent dyes, and many far-red dyes are used for SRRF imaging. Far-red dyes have low optical scattering, making them suitable for deep tissue and in vivo imaging. Nitrogen-vacancy (NV) defects in nanodiamonds can emit fluorescence under two-photon excitation, enabling imaging in deeper tissues with better signal-to-noise ratios. Moreover, optical distortions caused by sample inhomogeneities can be corrected using adaptive optics, enhancing imaging quality. Johnstone GE et al. used a custom two-photon microscope, combining adaptive optics and super-resolution radial fluctuation (SRRF) algorithms to perform two-photon excitation imaging of nanodiamonds and process the resulting images and achieved super-resolution imaging. Silver nanoclusters (AgNCs) have excellent biocompatibility, ultrafine size, and tunable optical and electronic properties, making them widely explored in biological applications. The emission wavelength of AgNCs falls in the red to near-infrared (NIR) region, about 50 nm redshifted from gold nanoclusters (AuNCs), making AgNCs ideal NIR probes for deep tissue penetration and minimal spontaneous fluorescence interference. Aditya Yadav et al. used bovine serum albumin-protected silver nanoclusters as specific fluorescent probes with infrared emission for super-resolution radial fluctuation imaging, achieving single-resolution imaging of lysosomes at approximately 126 nm. This opens up possibilities for specific labeling of live biological organelles and real-time event capture. 29 8.4 | Combination with other imaging techniques One of the key advantages of SRRF is its potential for integration with other imaging techniques. In addition to conventional wide-field fluorescence microscopy, total internal reflection fluorescence microscopy, and laser scanning confocal microscopy, SRRF is also highly compatible with electron microscopy, expansion microscopy, electrochemiluminescence microscopy, and ultrasound imaging. This flexibility allows it to handle different types of image data, which is crucial for gaining a deeper understanding of the ultrastructure and functions of biological systems. 8.4.1 | Correlative optoelectronic imaging Rescently researchers have developed various new optical-electron correlative imaging techniques (Correlative Light and Electron Microscopy, CLEM), which combine the specificity and biocompatibility of optical microscopy with the high resolution of electron microscopy. Dorothea Pinotsi et al. integrated a commercial wide-field fluorescence microscope into the sample chamber of a scanning electron microscope (SEM) and combined it with the SRRF algorithm, achieving correlative imaging of super-resolution fluorescence microscopy and scanning electron microscopy. This approach provided more detailed information for studying structures such as protein fibers. 12 8.4.2 | ExM-SRRF Expansion Microscopy (ExM) enables the physical expansion of samples, allowing details that are otherwise undetectable under traditional optical microscopy to be clearly observed. Combining SRRF with ExM achieves a resolution of up to 25 nm in an LED-based wide-field microscope system. This technology offers a simple, robust, scalable, and accessible tool for nanometer-scale molecular analysis in clinical pathology. Kylies D et al. used ExM-SRRF imaging to achieve super-resolution imaging of mitochondrial structures in human kidney tissue, cytoskeletal disorganization in human glioblastoma, and microvascular β-amyloid deposition in the brains of Alzheimer’s disease patients. 13 8.4.3 | Electrochemiluminescence microscopy (ECLM) Electrochemiluminescence Microscopy (ECLM) is an emerging imaging technique that combines electrochemical excitation with optical readout, offering advantages in high spatiotemporal resolution. However, traditional ECLM is limited by the optical diffraction limit and cannot achieve sub-200 nm resolution. Chen MM et al. first proposed super-resolution ECLM, using the SRRF algorithm to achieve approximately 100-nm spatial resolution. This technique enabled the measurement of site- and facet-specific activities of individual nanoparticles. The method revealed surface activity dependent on the facets and defects of individual nanoparticles, as well as the dynamic fluctuations of reaction activity patterns, providing new perspectives for catalysis, biological imaging, and single-entity analysis. 14 8.4.4 | Ultrasound diffraction attenuation microscopy (UDAM) Ultrasound imaging technology is widely used in clinical diagnosis due to its non-invasive nature and real-time capabilities. However, traditional ultrasound imaging is limited by the acoustic diffraction limit, making it difficult to resolve small vascular structures. The emergence of Super-Resolution Ultrasound (SRUS) technology has made it possible to exceed the acoustic diffraction limit, significantly improving imaging spatial resolution. Zhang J et al. developed UDAM, which replaces microbubble localization with super-resolution radial fluctuation to overcome the acoustic diffraction limit. This method provides a non-invasive tool for in vivo super-resolution microvascular imaging, which has significant implications for clinical diagnosis. 15 | Conclusion Here we explored the basic principles, workflow, methods, quality evaluation, advantages and disadvantages, optimization techniques, and applications of SRRF. With its high spatial resolution and relatively simple implementation, SRRF has overcome the resolution limitations of traditional fluorescence microscopy and become an important tool in biomedical research. However, when analyzing the findings from different studies on SRRF, it is important to recognize that some controversies still exist in this field. For example, while SRRF offers significant advantages in resolution enhancement, the complexity of data processing and the sample preparation requirements may impact the reliability of the results. Therefore, future research should focus on optimizing standardized processes for SRRF to ensure the reproducibility and comparability of results across different laboratories. In their future, the SRRF development is likely to focus on several key fields. First, with advancements in computational technology, the integration of artificial intelligence and machine learning methods is expected to lead to greater breakthroughs in data processing and image reconstruction, further enhancing SRRF’s potential for studying dynamic biological processes. Second, multimodal imaging will help overcome the limitations of SRRF in certain specific applications, thus expanding its range of use. Finally, as optical component manufacturing technologies advance, the development of higher-performance microscopes and optical systems will provide better hardware support for SRRF applications. References： 1. Valli J, Sanderson J. Super-Resolution Fluorescence Microscopy Methods for Assessing Mouse Biology [J]. Curr Protoc, 2021,1(8):e224. 2. Klar TA, Hell SW. Subdiffraction resolution in far-field fluorescence microscopy [J]. Opt Lett. 1999 Jul 15;24(14):954-960 3. Patterson G, Davidson M, Manley S, Lippincott-Schwartz J. Superresolution imaging using single-molecule localization [J]. Annu Rev Phys Chem, 2010, 61:345-367. 4. Gustafsson MG. Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy [J]. J Microsc, 2000, 198(2):82-87 5. Mukamel EA, Babcock H, Zhuang X. Statistical deconvolution for superresolution fluorescence microscopy [J]. Biophys J, 2012, 102(10):2391-2400. 6. Rosten E, Jones GE, Cox S. ImageJ plug-in for Bayesian analysis of blinking and bleaching. Nat Methods, 2013, 10(2):97-98 7. Dertinger T, Colyer R, Iyer Get et al .. Fast, background-free,3D super-resolution optical fluctuation imaging (SOFI) [J]. Proc Natl Acad Sci U S A, 2009, 106(52):22287-22292. 8. Gustafsson N, Culley S, Ashdown G, et al .. Fast live-cell conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations [J]. Nat Commun, 2016, 7:12471 9. Salsman J, Dellaire G. Super-Resolution Radial Fluctuations (SRRF) Microscopy [J]. Methods Mol Biol, 2022, 2440:225-251. 10. Culley S, Albrecht D, Jacobs C, et al .. Quantitative mapping and minimization of super-resolution optical imaging artifacts [J]. Nat Methods, 2018, 15(4):263-266. 11. Culley S, Tosheva KL, Matos Pereira P, et al .. SRRF: Universal live-cell super-resolution microscopy [J]. Int J Biochem Cell Biol, 2018, 101:74-79. 12. Pinotsi D, Rodighiero S, Campioni S, et al .. An Easy Path for Correlative Electron and Super-Resolution Light Microscopy [J]. Sci Rep, 2019, 9(1):15526. 13. Kylies D, Zimmermann M, Haas F, et al .. Expansion-enhanced super-resolution radial fluctuations enable nanoscale molecular profiling of pathology specimens [J]. Nat Nanotechnol, 2023, 18(4):336-342. 14. Chen MM, Xu CH, Zhao W, et al .. Super-Resolution Electrogenerated Chemiluminescence Microscopy for Single-Nanocatalyst Imaging [J]. J Am Chem Soc, 2021, 143(44):18511-18518. 15. Zhang J, Li N, Dong F, Liang S, et al .. Ultrasound Microvascular Imaging Based on Super-Resolution Radial Fluctuations [J]. J Ultrasound Med, 2020, 39(8):1507-1516. 16. Chen R, Zhao Y, Li M, et al .. Efficient super-resolution volumetric imaging by radial fluctuation Bayesian analysis light-sheet microscopy [J]. J Biophotonics, 2020, 13(8):e201960242 . 17. Han Y, Lu X, Zhang Z, et al .. Ultra-fast, universal super-resolution radial fluctuations (SRRF) algorithm for live-cell super-resolution microscopy [J]. Opt Express, 2019, 27(26):38337-38348. 18. Gong X, Zhou L, Yao L, et al .. Achieving increased resolution and reconstructed image quality with gradient variance modified super-resolution radial fluctuations [J]. ACS Photonics, 2022, 9(5):1700-1708. 19. Laine RF, Heil HS, Coelho S, et al .. High-fidelity 3D live-cell nanoscopy through data-driven enhanced super-resolution radial fluctuation [J]. Nat Methods, 2023, 20(12):1949-1956. 20. Li Y, Liu L, Roberts SK, Wang L. Super-resolution radial fluctuations microscopy for optimal resolution and fidelity [J]. Opt Lett, 2024, 49(10):2621-2624. 21. Castillo-Badillo JA, Bandi AC, Harlalka S, et al .. SRRF-Stream Imaging of Optogenetically Controlled Furrow Formation Shows Localized and Coordinated Endocytosis and Exocytosis Mediating Membrane Remodeling [J]. ACS Synth Biol, 2020, 9(4):902-919. 22. Chen, R. Tang, X. Zhao, Y. et al .. Single-Frame Deep-Learning Super-ResolutionMicroscopy for Intracellular Dynamics Imaging [J]. Nat. Commun, 2023, 14 (1):2854. 23. Anjum F, Kaushik K, Salam A, et al .. Super-Resolution Microscopy Unveils Synergistic Structural Changes of Organelles Upon Point Mutation [J]. Adv Biol (Weinh), 2024, 8(3):e2300399. 24. Gilleron J, Chafik A, Lacas-Gervais S, et al .. Golgi-associated retrograde protein (GARP) complex-dependent endosomes to trans Golgi network retrograde trafficking is controlled by Rab4b [J]. Cell Mol Biol Lett, 2024, 29(1):54. 25. Donaldson LA. Super-resolution imaging of Douglas fir xylem cell wall nanostructure using SRRF microscopy [J]. Plant Methods, 2022, 18(1):27. 26. Lee YH, Zhang S, Mitchell CK, et al .. Calcium Imaging with Super-Resolution Radial Fluctuations [J]. Biosci Bioeng, 2018, 4(4):78-84. 27. Venkatachalapathy M, Belapurkar V, Jose M, et al .. Live cell super resolution imaging by radial fluctuations using fluorogen binding tags [J]. Nanoscale, 2019, 11(8):3626-3632. 28. Johnstone GE, Cairns GS, Patton BR. Nanodiamonds enable adaptive-optics enhanced, super-resolution, two-photon excitation microscopy [J]. R Soc Open Sci, 2019, 6(7):190589. 29. Aditya Y, Kush K, Shagun S, et al .. Near-Infrared-Emitting Silver Nanoclusters as Fluorescent Probes for Super-resolution Radial Fluctuation Imaging of Lysosomes [J]. 2022, 5(7) :9260-9265 30. Garg R, Anjum F, Salam A, et al .. Tracking the super resolved structure of mitochondria using red emissive carbon nanodots as a fluorescent biomarker [J]. Chem Commun (Camb), 2023, 59(90):13454-13457. Legends Figure 1 | Principle and workflow of SRRF. A.Image acquisition. Time series images are acquired at short intervals using a conventional wide-field fluorescence microscope, total internal reflection fluorescence microscope, or laser confocal microscope; B. Spatial analysis; Pixels are divided into subpixels. Pixels of signal have a high degree of convergence (being the center of the PSF), pixels of background have little to no degree of convergence, due to background lacking radial symmetry; C. Temporal analysis; By tracking the radiality over time, it is clear if a subpixel contains a temporary signal (peak of radiality), or noise (unchanging low radiality); D:Super-Resolution Image Reconstruction. Using iterative algorithms, SRRF reconstructs a high-resolution image by combining radiality metrics, temporal correlation maps, and raw intensity data.Partial Image from Culley et al. 2018 . Figure 2 | SRRF used for super-resolution imaging of tubulin, Golgi, mitochondria and nuclear pore complex （NPC) in Hela cells (A-B) Spinning disk confocal microscope (SDCM) versus SRRF images (SRRF) of tubulin labelled by abberior LIVE 610 dye. (C-D) SDCM versus SRRF images of Golgi labelled by giantin immunofluorescence. (E-F) SDCM versus SRRF images of mitochondria labelled by Abberior Live Orange Mito dye (G-H)SDCM versus SRRF images of NPC labelled by Nup 98 immunofluorescence. Scale bars, 3 µm. Figure 3 | SRRF used for super-resolution live-imaging of Jurkat T cells transiently expressing LifeAct-GFP. (a) Conventional TIRF microscopy (TIRF) versus SRRF images (SRRF) of Jurkat T cells transfected with LifeAct-GFP and imaged for 180 s at 1 superresolution f.p.s. (b) Conventional TIRF and SRRF images of Jurkat T cells expressing LifeAct-GFP imaged on coverslips coated with anti-CD28 alone, antiCD3 alone or in combination (anti-CD3 and -CD28) to stimulate an immunological synapse formation (highlighted area corresponds to the region used for PIV analysis). (c) PIV analysis shows notable retrograde actin flow in anti-CD3 but not in anti-CD28 stimulated Jurkat T cells. A colour-coded measure of flow directionality and speed is plotted for the blue highlighted regions. White arrows in colour wheel are representative of flow direction, central colour (minimum intensity) corresponds to 0 mm min 1, saturated colours (maximum intensity) correspond to 38.4 mm min 1. Scale bars, 5 mm.Image from Nils Gustafsson et al. 2016. Table 1 | Comparison of SRRF with SMLM, STED and SIM in terms of hardware requirements, switchable fluorescent molecules, resolution, imaging speed, phototoxicity and live-cell compatibility. Terms SMLM STED SIM SRRF Hardware requirements High very high High Low Switchable fluorescent molecules Yes no no no Resolution (nm) 20-40 30-80 80-130 50-100 Imaging speed(Frame/s) 0.001 0.05-0.3 0.1-10 1 Phototoxicity High very high Middle Low Live-cell compatibility * ** *** **** Artifact occurrence likelihood ** * **** *** Information & Authors Information Version history V1 Version 1 17 March 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords advances applications principle super-resolution radial fluctuations (srrf) Authors Affiliations Sanhua Fang 0000-0003-0525-642X [email protected] Zhejiang University View all articles by this author Li Liu Zhejiang University View all articles by this author Dan Yang Zhejiang University View all articles by this author Shuangshuang Liu Zhejiang University View all articles by this author Qiong Huang Zhejiang University View all articles by this author Metrics & Citations Metrics Article Usage 401 views 222 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Sanhua Fang, Li Liu, Dan Yang, et al. Super-Resolution Radial Fluctuations (SRRF): A Versatile and Accessible Tool for Live-Cell Nanoscopy and Multimodal Imaging. Authorea . 17 March 2025. DOI: https://doi.org/10.22541/au.174224393.37284045/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174224393.37284045/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a00080cf1f73df88',t:'MTc3OTUwMjAwNQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();