Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Overview Of Cell Separation And Isolation01:20

Overview Of Cell Separation And Isolation

5.7K
Cell separation was first achieved in 1964 by S. H. Seal, who separated large tumor cells from the smaller blood cells using filtration. Two years later, Pohl and Hawk performed experiments on how cells respond differently to a nonuniform electric field based on the cell type. Such observations were the inception of cell separation methods, which allow isolating a single cell type from a heterogeneous sample.
5.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Modeling Vehicle Dust Extraction Impeller Degradation Using TOPSIS-Selected Optimal Degradation Trajectory.

Materials (Basel, Switzerland)·2026
Same author

Crosstalk in Alzheimer's-delirium nexus: Molecular mechanisms and therapeutic repurposing.

International review of neurobiology·2026
Same author

MedNLP-Hub: A knowledgebase platform for biomedical NLP tool discovery in clinical informatics.

International journal of medical informatics·2026
Same author

Dendritic Cell α-Ketoglutarate Regulates Tfh Polarization in Allergy.

Allergy·2026
Same author

Role of Phenolic Nanocompounds in Inflammatory Disorders: Current View and Future Aspects.

Current pharmaceutical design·2026
Same author

Innovative diabetes mellitus treatment strategies: Mesenchymal stem cell-based therapy and its impact on pro- and anti-inflammatory cytokines modulation.

Journal of pharmaceutical analysis·2026

Related Experiment Video

Updated: Jul 5, 2025

Single-cell RNA-Seq of Defined Subsets of Retinal Ganglion Cells
11:26

Single-cell RNA-Seq of Defined Subsets of Retinal Ganglion Cells

Published on: May 22, 2017

13.8K

scFed: federated learning for cell type classification with scRNA-seq.

Shuang Wang1,2, Bochen Shen2, Lanting Guo2

  • 1Joint Laboratory of Artificial Intelligence for Critical Care Medicine, Department of Critical Care Medicine and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, 610212, Chengdu, China.

Briefings in Bioinformatics
|January 15, 2024
PubMed
Summary

Federated learning, using the scFed framework, enables efficient and private cell identification from single-cell RNA sequencing data. It achieves competitive accuracy across diverse datasets, highlighting its potential for collaborative biomedical research.

Keywords:
cell typeclassificationfederated learningscRNA-seq

More Related Videos

Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues
10:12

Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues

Published on: January 10, 2019

18.5K
Isolation and Transcriptome Analysis of Plant Cell Types
08:53

Isolation and Transcriptome Analysis of Plant Cell Types

Published on: April 7, 2023

1.5K

Related Experiment Videos

Last Updated: Jul 5, 2025

Single-cell RNA-Seq of Defined Subsets of Retinal Ganglion Cells
11:26

Single-cell RNA-Seq of Defined Subsets of Retinal Ganglion Cells

Published on: May 22, 2017

13.8K
Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues
10:12

Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues

Published on: January 10, 2019

18.5K
Isolation and Transcriptome Analysis of Plant Cell Types
08:53

Isolation and Transcriptome Analysis of Plant Cell Types

Published on: April 7, 2023

1.5K

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Single-cell RNA sequencing (scRNA-seq) reveals cellular heterogeneity but generates large, sparse datasets. Data privacy regulations pose challenges for centralized analysis and cell identification. Federated learning offers a privacy-preserving approach for distributed data analysis.
  • Existing cell identification methods struggle with the scale and privacy constraints of scRNA-seq data.
  • Federated learning enables collaborative analysis of sensitive biological data without direct data sharing.

Purpose of the Study:

  • To introduce scFed, a unified federated learning framework for benchmarking cell identification algorithms on scRNA-seq data.
  • To evaluate the performance of different classification algorithms within a federated learning setting.
  • To assess the privacy-preserving capabilities and efficiency of federated learning for scRNA-seq data analysis.

Main Methods:

  • Developed scFed, a federated learning framework for scRNA-seq data analysis.
  • Benchmarked four classification algorithms (single-cell-specific and general-purpose) using scFed.
  • Evaluated performance on eight diverse, publicly available scRNA-seq datasets.
  • Conducted intra-dataset and inter-dataset experiments to assess model generalizability.

Main Results:

  • scFed demonstrated competitive accuracy compared to centralized models across various scRNA-seq datasets.
  • Transformer-based models showed strong performance in centralized settings but lagged slightly within the scFed framework.
  • Single-cell-specific models performed well within scFed, with Transformer models exhibiting higher time complexity.
  • The framework successfully enabled privacy-preserving benchmarking of cell identification algorithms.

Conclusions:

  • Federated learning, via scFed, is a viable and effective approach for privacy-preserving cell identification in scRNA-seq data.
  • scFed facilitates the selection of appropriate cell identification methods for federated analysis.
  • The study underscores the potential of federated learning for secure, collaborative biomedical research and data analysis.