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 Video

Updated: Jul 15, 2026

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

scDifformer: diffusion-based post-training for virtual cell modeling across large-scale single-cell data.

Zhan Xiao¹, Wuke Wang², Xin Long^1,3

¹Research Center for Life Sciences Computing, Zhejiang Lab, Hangzhou, Zhejiang 311121, China.

Nucleic Acids Research

|July 14, 2026

Summary

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

Transcriptome study of <i>Talaromyces verruculosus</i> TS63-9 for pectin-degrading CAZymes.

Microbiology spectrum·2026

Same author

Lamprey 3D single-cell transcriptomics reveals ancestral and specialized features of the vertebrate brain.

Science (New York, N.Y.)·2026

Same author

Guide RNA reprogramming facilitates minimized tracrRNA-dependent off-target and versatile CRISPR/Cas9 engineering.

Nature communications·2026

Same author

Optimization of IS621 recombinase/bridge RNA-directed recombination for precise insertion of large DNA fragments in human cells.

Nature communications·2026

Same author

Uncovering functional variants using a high-efficiency PE3-based screening platform.

Science advances·2026

Same author

ANXA11 suppression restores muscular function in the mdx mouse model of Duchenne muscular dystrophy (DMD).

Nature communications·2026

scDifformer, a novel AI model, enhances virtual cell development by accurately modeling complex single-cell data. It effectively denoises noisy biological information, improving cell type identification and analysis across diverse studies.

Area of Science:

Computational Biology
Bioinformatics
Artificial Intelligence in Biology

Background:

Virtual cells offer a powerful approach to understanding cellular processes.
Accurate modeling of single-cell data is crucial for virtual cell development.
Technical noise and biological heterogeneity present significant challenges in single-cell data analysis.

Purpose of the Study:

To develop a robust computational framework for modeling heterogeneous single-cell data.
To enhance the accuracy and generalizability of virtual cell models.
To address noise and batch effects in multi-modal single-cell datasets.

Main Methods:

Introduction of scDifformer, a transformer model with a denoising diffusion module.
A three-phase design: masked language model pre-training, diffusion-driven post-training, and fine-tuning.

More Related Videos

Analysis of Multidimensional Microscopy Data Using Cell-ACDC

Analysis of Multidimensional Microscopy Data Using Cell-ACDC

Published on: November 7, 2025

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

Related Experiment Videos

Last Updated: Jul 15, 2026

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

Analysis of Multidimensional Microscopy Data Using Cell-ACDC

Analysis of Multidimensional Microscopy Data Using Cell-ACDC

Published on: November 7, 2025

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

Integration with graph neural networks for spatial transcriptomics analysis.

Main Results:

scDifformer demonstrates improved cross-dataset performance and noise reduction.
Achieved state-of-the-art cell type annotation across diverse tissues and studies.
Successfully resolved immune cell identities and identified key biological markers and pathways.
Enhanced deconvolution accuracy in spatial transcriptomics data.

Conclusions:

scDifformer provides a scalable and biologically grounded framework for virtual cell modeling.
The model effectively handles noisy and heterogeneous single-cell data.
Offers a foundation for high-fidelity, multi-modal virtual cell development.