Recent development of computational cluster analysis methods for single-molecule localization microscopy images

Hyun, Yoonsuk; Kim, Doory

doi:10.1016/j.csbj.2023.01.006

Cited by 10 publications

(8 citation statements)

References 53 publications

(72 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although we here developed the silica- or silicon-specific super-resolution imaging method, we expect that our development would be extended to the development of the super-resolution imaging method for other materials in the future, allowing the investigation of nanostructures and the detection of nanodefects in semiconductor materials with various compositions. Additionally, with the development of a fast camera with a relatively high frame rate and the fast single-molecule localization algorithm using deep learning, we anticipate that this method will yield significantly high-throughput metrology. , As the dye labeling process (∼1 h) can be performed for the entire whole wafer sample area at one time no matter how large the wafer is, the entire dye-labeled wafer sample is expected to be examined with high-throughput when automatic mosaic STORM imaging is combined with the use of a high frame rate camera . Given its high labeling density, high chemical specificity, single-molecule sensitivity, and nanoscale resolution, our new fluorophore labeling and imaging approach for inorganic nanomaterials is ideal for the functional characterization of nanopatterns in silicon wafers, driving further innovation of metrology tools and applications.…”

Section: Discussionmentioning

confidence: 99%

Development of Highly Dense Material-Specific Fluorophore Labeling Method on Silicon-Based Semiconductor Materials for Three-Dimensional Multicolor Super-Resolution Fluorescence Imaging

Jeong

et al. 2023

Chem. Mater.

Self Cite

View full text Add to dashboard Cite

The recent development of super-resolution fluorescence microscopy (SRM) has drastically improved the resolution of light microscopy to the order of tens of nanometers. However, the application of SRM to semiconductor materials remains challenging because fluorophore labeling on inorganic materials with a high labeling density required for nanoimaging has been limited with conventional surface functionalization methods. Here, a novel approach for highly dense material-specific fluorophore labeling methods on silicon-based materials has been developed and demonstrated for SRM imaging of semiconductor line patterns. This approach is shown to selectively and sensitively probe different-sized silicon and silica line patterned arrays including edge structures on a wafer in three dimension, which has not been resolved by a conventional metrology system. Furthermore, we successfully demonstrate that this new method can detect nanoparticle defects with high sensitivity, suggesting its capability as an inspection tool for semiconductor defects. This new nanomaterial imaging approach is expected to drive further innovations in metrology tools and applications.

show abstract

Section: Discussionmentioning

confidence: 99%

Development of Highly Dense Material-Specific Fluorophore Labeling Method on Silicon-Based Semiconductor Materials for Three-Dimensional Multicolor Super-Resolution Fluorescence Imaging

Jeong

et al. 2023

Chem. Mater.

Self Cite

View full text Add to dashboard Cite

show abstract

“…While software packages for reconstruction and visualization of SMLM data are well-developed, [11] approaches for quantitative analysis of 2D or 3D point cloud data remain limited. Clustering analysis methods for SMLM include statistical, Bayesian, density-based, correlation-based, tessellation-based, image-based, and machine-learning based approaches [12, 13] . Previously, we applied batch SuperResNET network analysis to SMLM data for caveolin-1 (CAV1) and identified caveolae and three distinct classes of non-caveolar scaffolds [14-16] .…”

Section: Introductionmentioning

confidence: 99%

SuperResNET GUI: model-free single molecule network analysis software achieves molecular resolution of Nup96

Li,

Khater,

Hallgrimson

et al. 2024

Preprint

View full text Add to dashboard Cite

SuperResNET is an integrated machine learning-based image analysis software for visualizing and quantifying 3D point cloud localization data acquired by single molecule localization microscopy (SMLM). The computational modules of SuperResNET include correction for multiple blinking of a single fluorophore, followed by denoising, segmentation (clustering), and feature extraction, which are then used for cluster group identification, modularity analysis, blob retrieval and visualization in 2D and 3D. Using publicly available dSTORM data, we demonstrate the potential of SuperResNET in network analysis of subcellular nucleoporin Nup96 structures, that present a highly organized octagon structure comprised of eight corners. While many nuclear pore structure are incomplete, SuperResNET effectively segments complete nuclear pores and Nup96 corners based on differential proximity threshold analysis. SuperResNET quantitatively analyzes features from segmented nuclear pore structures, including complete structures with 8-fold symmetry, and from segmented corners. Application of SuperResNET network modularity module to segmented corners distinguishes two modules at 11.1 nm distance, corresponding to two individual Nup96 molecules. SuperResNET network analysis of SMLM data is therefore a model-free tool that can reconstruct network architecture and molecular distribution of subcellular structures without the bias of a specified prior model, attaining molecular resolution from dSTORM data. SuperResNET provides the user with flexibility to report on structural diversity in situ within the cell without model-fitting, providing opportunities for biological discovery.

show abstract

“…REPLOM utilizes the kinetic behaviour of self-assembly systems and photobleaching by surface docking in a photo-unstable environment, unlocking the temporal resolution of self-assembly kinetic pathways. Despite progress in acquiring these information-rich data sets, the analysis and identification of individual protein assemblies in SMLM are often reliant on manual annotations or system-specific approaches which are resource and time-strenuous and lack generalisation [29][30][31] .…”

mentioning

confidence: 99%

mentioning

confidence: 99%

“…The advancement of machine learning-based approaches [32][33][34][35][36][37][38][39][40][41][42] has been instrumental for quantitative image analysis 30,43,44 and has the potential to resolve the bottleneck of extraction of assemblies of interest in super-resolution data. In general, these approaches can be broadly categorised as either supervised or unsupervised, each of which has advantages and limitations 30 . Supervised algorithms are highly accurate when large amounts of annotated data are available albeit annotations require extensive manual labor and expert knowledge and the resulting model is often suitable for one specific data set or task.…”

mentioning

confidence: 99%

See 1 more Smart Citation

SEMORE: SEgmentation and MORphological fingErprinting by machine learning automates super-resolution data analysis

Bender,

Dreisler,

Zhang

et al. 2024

Nat Commun

View full text Add to dashboard Cite

The morphology of protein assemblies impacts their behaviour and contributes to beneficial and aberrant cellular responses. While single-molecule localization microscopy provides the required spatial resolution to investigate these assemblies, the lack of universal robust analytical tools to extract and quantify underlying structures limits this powerful technique. Here we present SEMORE, a semi-automatic machine learning framework for universal, system- and input-dependent, analysis of super-resolution data. SEMORE implements a multi-layered density-based clustering module to dissect biological assemblies and a morphology fingerprinting module for quantification by multiple geometric and kinetics-based descriptors. We demonstrate SEMORE on simulations and diverse raw super-resolution data: time-resolved insulin aggregates, and published data of dSTORM imaging of nuclear pore complexes, fibroblast growth receptor 1, sptPALM of Syntaxin 1a and dynamic live-cell PALM of ryanodine receptors. SEMORE extracts and quantifies all protein assemblies, their temporal morphology evolution and provides quantitative insights, e.g. classification of heterogeneous insulin aggregation pathways and NPC geometry in minutes. SEMORE is a general analysis platform for super-resolution data, and being a time-aware framework can also support the rise of 4D super-resolution data.

show abstract

Recent development of computational cluster analysis methods for single-molecule localization microscopy images

Cited by 10 publications

References 53 publications

Development of Highly Dense Material-Specific Fluorophore Labeling Method on Silicon-Based Semiconductor Materials for Three-Dimensional Multicolor Super-Resolution Fluorescence Imaging

Development of Highly Dense Material-Specific Fluorophore Labeling Method on Silicon-Based Semiconductor Materials for Three-Dimensional Multicolor Super-Resolution Fluorescence Imaging

SuperResNET GUI: model-free single molecule network analysis software achieves molecular resolution of Nup96

SEMORE: SEgmentation and MORphological fingErprinting by machine learning automates super-resolution data analysis

Contact Info

Product

Resources

About