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 Experiment Video

Updated: Jul 12, 2026

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
08:59

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

Published on: October 28, 2018

Multi-modality Graph Representation Learning for Malignant Cell Identification from scRNA-seq using DeepMalignant.

Pankaj Bhattarai, Weiman Yuan, Hongmei Chi

    Biorxiv : the Preprint Server for Biology
    |July 10, 2026
    PubMed
    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

    Benchmarking cell type annotation in spatial transcriptomics: resolving cellular hierarchies, biological fidelity, and dynamic cell states.

    bioRxiv : the preprint server for biology·2026
    Same author

    Interpretable spatial multi-omics data integration and dimensionality reduction with SpaMV.

    Nature communications·2026
    Same author

    CSsingle: a unified tool for robust decomposition of bulk and spatial transcriptomic data across diverse single-cell references.

    Nucleic acids research·2026
    Same author

    A little longer, a lot better: simulation-guided exploration of extended-length single-end barcoded reads for structural variant detection.

    bioRxiv : the preprint server for biology·2026
    Same author

    scLongTree: an accurate computational tool to infer the longitudinal tree for scDNAseq data.

    bioRxiv : the preprint server for biology·2026
    Same author

    Trajectories of Response Inhibition Development in Adolescence.

    bioRxiv : the preprint server for biology·2026

    DeepMalignant accurately identifies malignant cells in single-cell RNA sequencing data by integrating gene expression and copy number alterations. This novel method improves upon existing tools for cancer genomics research.

    Area of Science:

    • Genomics
    • Computational Biology
    • Cancer Research

    Background:

    • Distinguishing malignant from normal cells in single-cell RNA sequencing (scRNA-seq) data is crucial for cancer genomics.
    • Current methods face challenges in precision, generalizability across cancer types, and robustness across sequencing platforms.

    Purpose of the Study:

    • To develop an advanced computational tool for unsupervised malignant cell identification.
    • To improve the accuracy and reliability of malignant cell classification in diverse cancer datasets.

    Main Methods:

    • Developed DeepMalignant, an unsupervised multimodal graph attention autoencoder.
    • Integrated gene expression and copy number alteration (CNA) data for malignant cell identification.
    • Applied and benchmarked DeepMalignant against state-of-the-art methods across multiple cancer types and sequencing platforms.

    Related Experiment Videos

    Last Updated: Jul 12, 2026

    Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
    08:59

    Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

    Published on: October 28, 2018

    Main Results:

    • DeepMalignant demonstrated superior performance in balancing precision and recall compared to existing methods.
    • The model consistently outperformed methods using only gene expression or CNA data, achieving higher F1 scores.
    • Ablation studies confirmed the independent contributions of CNA-based edge weighting and graph attention aggregation to performance.
    • Attribution analysis revealed that learned embeddings capture biologically relevant malignant programs.
    • Application to ductal carcinoma in situ (DCIS) samples showed high consistency between DeepMalignant-identified tumor regions and histological images.
    • Cell-cell communication analysis revealed differential signaling from stromal cells to normal versus malignant epithelial cells.

    Conclusions:

    • DeepMalignant offers a robust and accurate solution for malignant cell identification in scRNA-seq data.
    • The integration of multimodal data (gene expression and CNA) enhances classification performance.
    • Accurate tumor-normal cell classification enables deeper biological insights into the tumor microenvironment and cell-cell interactions.