stDyer enables spatial domain clustering with dynamic graph embedding

Abstract Spatially resolved transcriptomics (SRT) data provide critical insights into gene expression patterns within tissue contexts, necessitating effective methods for identifying spatial domains. We introduce stDyer, an end-to-end deep learning framework for spatial domain clustering in SRT data...

Full description

Saved in:
Bibliographic Details
Main Authors: Ke Xu, Yu Xu, Zirui Wang, Xin Maizie Zhou, Lu Zhang
Format: Article
Language:English
Published: BMC 2025-02-01
Series:Genome Biology
Subjects:
Online Access:https://doi.org/10.1186/s13059-025-03503-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849724004690034688
author Ke Xu
Yu Xu
Zirui Wang
Xin Maizie Zhou
Lu Zhang
author_facet Ke Xu
Yu Xu
Zirui Wang
Xin Maizie Zhou
Lu Zhang
author_sort Ke Xu
collection DOAJ
description Abstract Spatially resolved transcriptomics (SRT) data provide critical insights into gene expression patterns within tissue contexts, necessitating effective methods for identifying spatial domains. We introduce stDyer, an end-to-end deep learning framework for spatial domain clustering in SRT data. stDyer combines Gaussian Mixture Variational AutoEncoder with graph attention networks to learn embeddings and perform clustering. Its dynamic graphs adaptively link units based on Gaussian Mixture assignments, improving clustering and producing smoother domain boundaries. stDyer’s mini-batch strategy and multi-GPU support facilitate scalability to large datasets. Benchmarking against state-of-the-art tools, stDyer demonstrates superior performance in spatial domain clustering, multi-slice analysis, and large-scale dataset handling.
format Article
id doaj-art-395ed433997f421faebd4c53621a7e1b
institution DOAJ
issn 1474-760X
language English
publishDate 2025-02-01
publisher BMC
record_format Article
series Genome Biology
spelling doaj-art-395ed433997f421faebd4c53621a7e1b2025-08-20T03:10:52ZengBMCGenome Biology1474-760X2025-02-0126112510.1186/s13059-025-03503-ystDyer enables spatial domain clustering with dynamic graph embeddingKe Xu0Yu Xu1Zirui Wang2Xin Maizie Zhou3Lu Zhang4Department of Computer Science, Hong Kong Baptist UniversityDepartment of Computer Science, Hong Kong Baptist UniversityDepartment of Computer Science, Hong Kong Baptist UniversityDepartment of Biomedical Engineering, Vanderbilt UniversityDepartment of Computer Science, Hong Kong Baptist UniversityAbstract Spatially resolved transcriptomics (SRT) data provide critical insights into gene expression patterns within tissue contexts, necessitating effective methods for identifying spatial domains. We introduce stDyer, an end-to-end deep learning framework for spatial domain clustering in SRT data. stDyer combines Gaussian Mixture Variational AutoEncoder with graph attention networks to learn embeddings and perform clustering. Its dynamic graphs adaptively link units based on Gaussian Mixture assignments, improving clustering and producing smoother domain boundaries. stDyer’s mini-batch strategy and multi-GPU support facilitate scalability to large datasets. Benchmarking against state-of-the-art tools, stDyer demonstrates superior performance in spatial domain clustering, multi-slice analysis, and large-scale dataset handling.https://doi.org/10.1186/s13059-025-03503-ySpatially resolved transcriptomicsSpatial domain clusteringDynamic graphsDeep learning
spellingShingle Ke Xu
Yu Xu
Zirui Wang
Xin Maizie Zhou
Lu Zhang
stDyer enables spatial domain clustering with dynamic graph embedding
Genome Biology
Spatially resolved transcriptomics
Spatial domain clustering
Dynamic graphs
Deep learning
title stDyer enables spatial domain clustering with dynamic graph embedding
title_full stDyer enables spatial domain clustering with dynamic graph embedding
title_fullStr stDyer enables spatial domain clustering with dynamic graph embedding
title_full_unstemmed stDyer enables spatial domain clustering with dynamic graph embedding
title_short stDyer enables spatial domain clustering with dynamic graph embedding
title_sort stdyer enables spatial domain clustering with dynamic graph embedding
topic Spatially resolved transcriptomics
Spatial domain clustering
Dynamic graphs
Deep learning
url https://doi.org/10.1186/s13059-025-03503-y
work_keys_str_mv AT kexu stdyerenablesspatialdomainclusteringwithdynamicgraphembedding
AT yuxu stdyerenablesspatialdomainclusteringwithdynamicgraphembedding
AT ziruiwang stdyerenablesspatialdomainclusteringwithdynamicgraphembedding
AT xinmaiziezhou stdyerenablesspatialdomainclusteringwithdynamicgraphembedding
AT luzhang stdyerenablesspatialdomainclusteringwithdynamicgraphembedding