Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

scMUSCL: multi-source transfer learning for clustering scRNA-seq data.

Arash Khoeini¹, Funda Sar², Yen-Yi Lin^2,3

¹School of Computing Science, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada.

Bioinformatics (Oxford, England)

|March 28, 2025

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by

Same author

The Effects of High-Thoracic Spinal Cord Injury on the Heart Transcriptome.

Journal of neurotrauma·2026

Same author

CAR-T Cell-Derived Exosomes and Cancer Immunotherapy: Advancing Production and Delivery Through Biofabrication.

Biofabrication·2026

Same author

Deep docking, part 2: an amplified DDU platform for ultra-large virtual screening.

Chemical science·2026

Same author

Microfluidic platforms for precision delivery of therapeutic cells in regenerative and personalized medicine.

Advanced drug delivery reviews·2026

Same author

Well-differentiated papillary mesothelial tumour: histologic, molecular and genetic features support a benign clonal neoplasm despite diffuse peritoneal involvement.

Histopathology·2025

Same author

Distinct transcription factor interactions drive HOXB13 activity in different stages of prostate cancer.

Proceedings of the National Academy of Sciences of the United States of America·2025

Same journal

conMItion: an R package adjusting confounding factors for associations in multi-omics.

Bioinformatics (Oxford, England)·2026

Same journal

SpaMFG: a Spatial Multi-omics Integration Method based on Feature Grouping.

Bioinformatics (Oxford, England)·2026

Same journal

CSCN: Inference of Cell-Specific Causal Networks Using Single-Cell RNA-Seq Data.

Bioinformatics (Oxford, England)·2026

Same journal

Sparse CCA-Based Mediation Analysis with High-Dimensional Exposures and Mediators.

Bioinformatics (Oxford, England)·2026

Same journal

Enhancing Cross-Context Generalization in Drug Perturbation Prediction with a Multimodal Conditional Diffusion Framework.

Bioinformatics (Oxford, England)·2026

Same journal

Primer Design through Submodular Function Estimation.

Bioinformatics (Oxford, England)·2026

See all related articles

This study introduces Single Cell MUlti-Source CLustering (scMUSCL), a novel transfer learning method for single-cell RNA sequencing (scRNA-seq) analysis. scMUSCL effectively identifies cell clusters by leveraging multiple annotated datasets, outperforming existing methods.

Area of Science:

Genomics
Bioinformatics
Computational Biology

Background:

Single-cell RNA sequencing (scRNA-seq) analysis requires robust clustering for downstream applications.
Unsupervised clustering methods often yield biologically irrelevant cell clusters due to the high dimensionality of single-cell data.
Existing methods overlook valuable annotated scRNA-seq datasets, limiting clustering performance.

Purpose of the Study:

To develop a novel transfer learning method, scMUSCL, for enhanced cell clustering in scRNA-seq data.
To leverage multiple annotated reference datasets to improve the biological relevance of cell clusters in target datasets.
To address discrepancies across datasets and eliminate the need for batch correction.

Main Methods:

scMUSCL utilizes a deep neural network for extracting domain- and batch-invariant cell representations.

Related Experiment Videos

The method integrates knowledge from multiple annotated source datasets to cluster target datasets.

scMUSCL does not require prior knowledge of the number of clusters in the target dataset.

Main Results:

scMUSCL consistently outperforms existing unsupervised and transfer learning-based clustering methods across 20 real-life datasets.
The method effectively benefits from multiple source datasets for improved clustering.
scMUSCL accurately estimates the number of clusters in the target dataset without prior specification.

Conclusions:

scMUSCL offers a significant advancement in scRNA-seq data analysis by improving clustering accuracy and biological interpretability.
The transfer learning approach effectively harmonizes data from diverse sources, enhancing downstream analyses.
This method provides a robust solution for identifying meaningful cell populations in complex scRNA-seq datasets.