DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads

The advent of single-cell sequencing has revolutionized the study of cellular dynamics, providing unprecedented resolution into the molecular states and heterogeneity of individual cells. However, the rich potential of exon-level information and junction reads within single cells remains underutiliz...

Full description

Saved in:
Bibliographic Details
Published in:Nature communications Vol. 16; no. 1; pp. 6202 - 26
Main Authors: Song, Kailu, Zheng, Yumin, Zhao, Bowen, Eidelman, David H., Tang, Jian, Ding, Jun
Format: Journal Article
Language:English
Published: London Nature Publishing Group UK 04.07.2025
Nature Publishing Group
Nature Portfolio
Subjects:
631/114/1305
631/114/1314
631/114/2397
631/114/794
Alternative splicing
Alternative Splicing - genetics
Animals
Aquatic mammals
Biomarkers
Clustering
Datasets
Deep Learning
Exons - genetics
Gene expression
Gene Expression Profiling - methods
Graph neural networks
Graph representations
Graphical representations
Heterogeneity
Humanities and Social Sciences
Humans
multidisciplinary
Science
Science (multidisciplinary)
Single-Cell Analysis - methods
Therapeutic targets
Transcriptome - genetics
Transcriptomics
ISSN:2041-1723, 2041-1723
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The advent of single-cell sequencing has revolutionized the study of cellular dynamics, providing unprecedented resolution into the molecular states and heterogeneity of individual cells. However, the rich potential of exon-level information and junction reads within single cells remains underutilized. Conventional gene-count methods overlook critical exon and junction data, limiting the quality of cell representation and downstream analyses such as subpopulation identification and alternative splicing detection. We introduce DOLPHIN, a deep learning method that integrates exon-level and junction read data, representing genes as graph structures. These graphs are processed by a variational graph autoencoder to improve cell embeddings. DOLPHIN not only demonstrates superior performance in cell clustering, biomarker discovery, and alternative splicing detection but also provides a distinct capability to detect subtle transcriptomic differences at the exon level that are often masked in gene-level analyses. By examining cellular dynamics with enhanced resolution, DOLPHIN provides new insights into disease mechanisms and potential therapeutic targets. Single-cell RNA-seq analysis is conventionally limited to gene-level quantification, missing transcript diversity. Here, authors present DOLPHIN, a deep learning method that enables exon- and junction-level analysis to improve cell representation and detect alternative splicing.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2041-1723
2041-1723
DOI:10.1038/s41467-025-61580-w