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: May 29, 2026

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

PVAED: prior-guided variational autoencoders with diffusion denoising for interpretable single-cell representation

Yawei Niu¹, Yichu Chen¹, Wenji Ma¹

¹Center for Single-Cell Omics, School of Public Health, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

Briefings in Bioinformatics

|May 27, 2026

Summary

This summary is machine-generated.

Related Concept Videos

Deconvolution

Deconvolution

Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...

You might also read

Related Articles

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

Sort by

Same author

Associations of serum per- and polyfluoroalkyl substances (PFAS) mixture with the risk of chronic diseases: evidence from single‑ , multi-pollutant, and machine learning models.

Archives of toxicology·2026

Same author

Paragonimiasis in a child with massive pericardial effusion as the primary manifestation: a case report.

Journal of cardiothoracic surgery·2026

Same author

SPHK1 deficiency promotes intestinal homeostasis by ameliorating ER stress-induced gastrointestinal injury during murine graft-versus-host disease.

Science China. Life sciences·2026

Same author

Label-free fluorescence lifetime imaging can distinguish cancer from healthy tissue in spontaneously occurring canine oral tumors.

Scientific reports·2026

Same author

Genetic variants in VEGF gene family enhance colorectal cancer susceptibility via modulating metabolic pathways.

Archives of toxicology·2025

Same author

Pharmacokinetics of lidocaine infusion during surgery for glioblastoma.

Clinical neurology and neurosurgery·2025

PVAED, a new dimensionality reduction framework, integrates biological knowledge into variational autoencoders (VAEs) to improve single-cell RNA sequencing analysis. It enhances cell representation and biological interpretability for identifying regulatory factors and cell subpopulations.

Area of Science:

Computational biology
Genomics
Bioinformatics

Background:

Single-cell RNA sequencing (scRNA-seq) analysis heavily relies on dimensionality reduction for pattern discovery.
Existing methods like UMAP and t-SNE struggle to capture regulatory information from gene expression data alone.
Identifying cellular pathways, protein complexes, and transcription factor targets requires deeper biological insights.

Purpose of the Study:

To develop a novel dimensionality reduction framework, PVAED, that integrates biological prior knowledge into variational autoencoders (VAEs).
To enhance the interpretability and performance of scRNA-seq data analysis by refining latent embeddings and preserving local cellular similarity.
To facilitate the identification of regulatory factors and biological insights from complex gene expression datasets.

Keywords:

VAE model diffusion model dimension reduction prior knowledge scRNA-seq

Related Experiment Videos

Last Updated: May 29, 2026

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

Main Methods:

PVAED framework combines a variational autoencoder (VAE) with biological prior knowledge integration.
A diffusion-based denoising module refines latent embeddings for improved data representation.
A neighborhood-preserving loss term ensures local structure similarity among cells in the reduced dimensional space.

Main Results:

PVAED demonstrated an average 43% improvement in low-dimensional cell representation compared to standard VAEs across seven clustering metrics.
Achieved 34% and 44% gains in global and local structure preservation, respectively, over classical and VAE-based methods.
PVAED successfully identified pre-committed neuronal subpopulations during mammalian cerebral cortex development.

Conclusions:

PVAED offers superior performance and interpretability for scRNA-seq data analysis by incorporating biological priors.
The framework aids in identifying critical regulatory factors linked to disease pathology and predicting related genes.
PVAED provides a powerful tool for uncovering complex biological processes and cell states from high-dimensional single-cell data.