Digital-SMLM for precisely localizing emitters within the diffraction limit DOI Creative Commons
Zhe Jia, Lingxiao Zhou, Haoyu Li

et al.

Nanophotonics, Journal Year: 2024, Volume and Issue: 13(19), P. 3647 - 3661

Published: June 5, 2024

Precisely pinpointing the positions of emitters within diffraction limit is crucial for quantitative analysis or molecular mechanism investigation in biomedical research but has remained challenging unless exploiting single molecule localization microscopy (SMLM). Via integrating experimental spot dataset with deep learning, we develop a new approach, Digital-SMLM, to accurately predict emitter numbers and sub-diffraction-limit spots an accuracy up 98 % root mean square error as low 14 nm. Digital-SMLM can resolve two at close distance, e.g. 30 outperforms Deep-STORM predicting sub-diffraction-limited recovering ground truth distribution molecules interest. We have validated generalization capability using independent data. Furthermore, complements SMLM by providing more accurate event number precise positions, enabling closely approximate natural state high-density cellular structures.

Language: Английский

Semiconducting polymer dots for multifunctional integrated nanomedicine carriers DOI Creative Commons
Ze Zhang, Chenhao Yu, Yuyang Wu

et al.

Materials Today Bio, Journal Year: 2024, Volume and Issue: 26, P. 101028 - 101028

Published: March 24, 2024

The expansion applications of semiconducting polymer dots (Pdots) among optical nanomaterial field have long posed a challenge for researchers, promoting their intelligent application in multifunctional nano-imaging systems and integrated nanomedicine carriers diagnosis treatment. Despite notable progress, several inadequacies still persist the Pdots, including development simplified near-infrared (NIR) nanoprobes, elucidation inherent biological behavior, integration information processing nanotechnology into biomedical applications. This review aims to comprehensively elucidate current status Pdots as classical nanophotonic material by discussing its advantages limitations terms biocompatibility, adaptability microenvironments vivo, etc. Multifunctional surface chemistry play crucial roles realizing Pdots. Information visualization based on physicochemical properties is pivotal achieving detection, sensing, labeling probes. Therefore, we refined underlying mechanisms constructed multiple comprehensive original mechanism summaries establish benchmark. Additionally, explored cross-linking interactions between nanomedicine, potential yet complete metabolic pathways, future research directions, innovative solutions integrating treatment strategies. presents possible expectations valuable insights advancing specifically from chemical, medical, photophysical practitioners' standpoints.

Language: Английский

Citations

8
AI analysis of super-resolution microscopy: Biological discovery in the absence of ground truth DOI Creative Commons
Ivan R. Nabi, Ben Cardoen, Ismail M. Khater

et al.

The Journal of Cell Biology, Journal Year: 2024, Volume and Issue: 223(8)

Published: June 12, 2024

Super-resolution microscopy, or nanoscopy, enables the use of fluorescent-based molecular localization tools to study structure at nanoscale level in intact cell, bridging mesoscale gap classical structural biology methodologies. Analysis super-resolution data by artificial intelligence (AI), such as machine learning, offers tremendous potential for discovery new biology, that, definition, is not known and lacks ground truth. Herein, we describe application weakly supervised paradigms microscopy its enable accelerated exploration architecture subcellular macromolecules organelles.

Language: Английский

Citations

5
A versatile automated pipeline for quantifying virus infectivity by label-free light microscopy and artificial intelligence DOI Creative Commons
Anthony Petkidis, Vardan Andriasyan, Luca Murer

et al.

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: June 15, 2024

Abstract Virus infectivity is traditionally determined by endpoint titration in cell cultures, and requires complex processing steps human annotation. Here we developed an artificial intelligence (AI)-powered automated framework for ready detection of virus-induced cytopathic effect (DVICE). DVICE uses the convolutional neural network EfficientNet-B0 transmitted light microscopy images infected including coronavirus, influenza virus, rhinovirus, herpes simplex vaccinia adenovirus. robustly measures effects (CPE), as shown class activation mapping. Leave-one-out cross-validation different types demonstrates high accuracy viruses, SARS-CoV-2 saliva. Strikingly, exhibits virus specificity, with adenovirus, herpesvirus, SARS-CoV-2. In sum, provides unbiased scores infectious agents causing CPE, can be adapted to laboratory diagnostics, drug screening, serum neutralization or clinical samples.

Language: Английский

Citations

5
Significance of Artificial Intelligence in the Study of Virus–Host Cell Interactions DOI Creative Commons
James Elste,

Akash Saini,

Rafael Mejía-Alvarez

et al.

Biomolecules, Journal Year: 2024, Volume and Issue: 14(8), P. 911 - 911

Published: July 26, 2024

A highly critical event in a virus's life cycle is successfully entering given host. This process begins when viral glycoprotein interacts with target cell receptor, which provides the molecular basis for virus-host interactions novel drug discovery. Over years, extensive research has been carried out field of interaction, generating massive number genetic and data sources. These datasets are an asset predicting at level using machine learning (ML), subset artificial intelligence (AI). In this direction, ML tools now being applied to recognize patterns these predict between virus host cells protein-protein protein-sugar levels, as well perform transcriptional translational analysis. On other end, deep (DL) algorithms-a subfield ML-can extract high-level features from very large hidden within genomic sequences images develop models rapid discovery predictions that address pathogenic viruses displaying heightened affinity receptor docking enhanced entry. DL pivotal forces, driving innovation their ability analysis enormous efficient, cost-effective, accurate, high-throughput manner. review focuses on complexity light current advances AI pathogenesis improve new treatments prevention strategies.

Language: Английский

Citations

4
Mantis: high-throughput 4D imaging and analysis of the molecular and physical architecture of cells DOI Creative Commons
Ivan E. Ivanov, Eduardo Hirata-Miyasaki, Talon Chandler

et al.

PNAS Nexus, Journal Year: 2024, Volume and Issue: 3(9)

Published: Aug. 9, 2024

High-throughput dynamic imaging of cells and organelles is essential for understanding complex cellular responses. We report Mantis, a high-throughput 4D microscope that integrates two complementary, gentle, live-cell technologies: remote-refocus label-free microscopy oblique light-sheet fluorescence microscopy. Additionally, we shrimPy (Smart Robust Imaging Measurement in Python), an open-source software imaging, deconvolution, single-cell phenotyping data. Using Mantis shrimPy, achieved high-content correlative molecular dynamics the physical architecture 20 cell lines every 15 min over 7.5 h. This platform also facilitated detailed measurements impacts viral infection on host proteins. The can enable profiling intracellular dynamics, long-term analysis responses to perturbations, optical screens dissect gene regulatory networks.

Language: Английский

Citations

2
Label-free microscopy for virus infections DOI Creative Commons
Anthony Petkidis, Vardan Andriasyan, Urs F. Greber

et al.

Microscopy, Journal Year: 2023, Volume and Issue: 72(3), P. 204 - 212

Published: April 20, 2023

Microscopy has been essential to elucidate micro- and nano-scale processes in space time provided insights into cell organismic functions. It is widely employed biology, microbiology, physiology, clinical sciences virology. While label-dependent microscopy, such as fluorescence provides molecular specificity, it remained difficult multiplex live samples. In contrast, label-free microscopy reports on overall features of the specimen at minimal perturbation. Here, we discuss modalities imaging molecular, cellular tissue levels, including transmitted light quantitative phase imaging, cryogenic electron or tomography atomic force microscopy. We highlight how used probe structural organization mechanical properties viruses, virus particles infected cells across a wide range spatial scales. working principles procedures analyses showcase they open new avenues Finally, orthogonal approaches that enhance complement techniques.

Language: Английский

Citations

6
Mantis: high-throughput 4D imaging and analysis of the molecular and physical architecture of cells DOI Creative Commons
Ivan E. Ivanov, Eduardo Hirata-Miyasaki, Talon Chandler

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Dec. 19, 2023

Abstract High-throughput dynamic imaging of cells and organelles is essential for understanding complex cellular responses. We report Mantis, a high-throughput 4D microscope that integrates two complementary, gentle, live-cell technologies: remote-refocus label-free microscopy oblique light-sheet fluorescence microscopy. Additionally, we shrimPy, an open-source software imaging, deconvolution, single-cell phenotyping data. Using Mantis achieved high-content correlative molecular dynamics the physical architecture 20 cell lines every 15 minutes over 7.5 hours. This platform also facilitated detailed measurements impacts viral infection on host proteins. The can enable profiling intracellular dynamics, long-term analysis responses to perturbations, optical screens dissect gene regulatory networks. Significance Statement Understanding interactions components crucial biological research drug discovery. Current methods only image few fluorescent labels, providing limited view these processes. developed 3D maps among systems. combines multiple fluorophores with quantitative complemented by our data acquisition high-performance analysis. enabled simultaneous time-lapse perturbations like at resolution. approach accelerate image-based

Language: Английский

Citations

5
Spatiotemporal visualization of DNA replication by click chemistry reveals bubbling of viral DNA in virion formation DOI Creative Commons
Alfonso Gómez-González, Patricia Burkhardt, Michael Bauer

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 16, 2024

Abstract The organisation of human chromosomes reversibly changes in cell division, and irreversibly apoptosis or erythropoiesis by DNA condensation fragmentation processes. Yet, how viral replication the nucleus affects host chromatin remains poorly understood. Here we used dual-color click chemistry to image adenovirus replication, demonstrating compaction during active expansion compartment (VRC). Early-replicated (vDNA) segregated from VRC lost phospho-serine5-RNA Pol-II DNA-binding protein (DBP), while late-replicated vDNA retained RNA Pol-II, besides RNA-splicing DNA-packaging proteins. Depending on assembly 52K, late-stage VRCs gave rise progeny droplet formation with GFP-tagged virion V into 52K biomolecular condensates. study reveals distinct functions early provides insight passive liquid phase separated zones conducive selective genome packaging nascent virions.

Language: Английский

Citations

1
Digital-SMLM for precisely localizing emitters within the diffraction limit DOI Creative Commons
Zhe Jia, Lingxiao Zhou, Haoyu Li

et al.

Nanophotonics, Journal Year: 2024, Volume and Issue: 13(19), P. 3647 - 3661

Published: June 5, 2024

Precisely pinpointing the positions of emitters within diffraction limit is crucial for quantitative analysis or molecular mechanism investigation in biomedical research but has remained challenging unless exploiting single molecule localization microscopy (SMLM). Via integrating experimental spot dataset with deep learning, we develop a new approach, Digital-SMLM, to accurately predict emitter numbers and sub-diffraction-limit spots an accuracy up 98 % root mean square error as low 14 nm. Digital-SMLM can resolve two at close distance, e.g. 30 outperforms Deep-STORM predicting sub-diffraction-limited recovering ground truth distribution molecules interest. We have validated generalization capability using independent data. Furthermore, complements SMLM by providing more accurate event number precise positions, enabling closely approximate natural state high-density cellular structures.

Language: Английский

Citations

0