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

Abstract 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 un...

Full description

Saved in:
Bibliographic Details
Main Authors: Kailu Song, Yumin Zheng, Bowen Zhao, David H. Eidelman, Jian Tang, Jun Ding
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-61580-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849402183401865216
author Kailu Song
Yumin Zheng
Bowen Zhao
David H. Eidelman
Jian Tang
Jun Ding
author_facet Kailu Song
Yumin Zheng
Bowen Zhao
David H. Eidelman
Jian Tang
Jun Ding
author_sort Kailu Song
collection DOAJ
description Abstract 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.
format Article
id doaj-art-3b6e719ce6e24e939f8aedfb86bb3cf0
institution Kabale University
issn 2041-1723
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-3b6e719ce6e24e939f8aedfb86bb3cf02025-08-20T03:37:37ZengNature PortfolioNature Communications2041-17232025-07-0116112610.1038/s41467-025-61580-wDOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction readsKailu Song0Yumin Zheng1Bowen Zhao2David H. Eidelman3Jian Tang4Jun Ding5Quantitative Life Sciences, McGill UniversityQuantitative Life Sciences, McGill UniversityMeakins-Christie Laboratories, Research Institute of the McGill University Health CentreMeakins-Christie Laboratories, Research Institute of the McGill University Health CentreHEC MontréalQuantitative Life Sciences, McGill UniversityAbstract 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.https://doi.org/10.1038/s41467-025-61580-w
spellingShingle Kailu Song
Yumin Zheng
Bowen Zhao
David H. Eidelman
Jian Tang
Jun Ding
DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
Nature Communications
title DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
title_full DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
title_fullStr DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
title_full_unstemmed DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
title_short DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
title_sort dolphin advances single cell transcriptomics beyond gene level by leveraging exon and junction reads
url https://doi.org/10.1038/s41467-025-61580-w
work_keys_str_mv AT kailusong dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads
AT yuminzheng dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads
AT bowenzhao dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads
AT davidheidelman dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads
AT jiantang dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads
AT junding dolphinadvancessinglecelltranscriptomicsbeyondgenelevelbyleveragingexonandjunctionreads