SpaIM: single-cell spatial transcriptomics imputation via style transfer

Abstract Spatial transcriptomics (ST) technologies have transformed our understanding of cellular organization but are limited by sparse signals and restricted gene coverage. To address these challenges, we introduce SpaIM, a style transfer learning model that leverages single-cell RNA sequencing (s...

Full description

Saved in:
Bibliographic Details
Main Authors: Bo Li, Ziyang Tang, Aishwarya Budhkar, Xiang Liu, Tonglin Zhang, Baijian Yang, Jing Su, Qianqian Song
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-63185-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849226151330840576
author Bo Li
Ziyang Tang
Aishwarya Budhkar
Xiang Liu
Tonglin Zhang
Baijian Yang
Jing Su
Qianqian Song
author_facet Bo Li
Ziyang Tang
Aishwarya Budhkar
Xiang Liu
Tonglin Zhang
Baijian Yang
Jing Su
Qianqian Song
author_sort Bo Li
collection DOAJ
description Abstract Spatial transcriptomics (ST) technologies have transformed our understanding of cellular organization but are limited by sparse signals and restricted gene coverage. To address these challenges, we introduce SpaIM, a style transfer learning model that leverages single-cell RNA sequencing (scRNA-seq) data to predict unmeasured gene expressions in ST profiles. By disentangling shared content and modality-specific styles, SpaIM effectively integrates scRNA-seq’s rich gene expression with the spatial context of ST. Evaluated across 53 datasets spanning sequencing- and imaging-based spatial technologies in various tissue types, SpaIM consistently outperforms 12 state-of-the-art methods in improving gene coverage and expression accuracy. Furthermore, SpaIM enhances downstream analyses, including ligand-receptor interaction inference, spatial domain characterization, and differential gene expression analysis. Released as open-source software, SpaIM expands accessibility and utility in ST research. Overall, SpaIM represents a robust and generalizable framework for enriching ST data with single-cell information, enabling deeper insights into tissue architecture and cellular function.
format Article
id doaj-art-9f8b3a990502487298f87bbf7a4b304f
institution Kabale University
issn 2041-1723
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-9f8b3a990502487298f87bbf7a4b304f2025-08-24T11:38:51ZengNature PortfolioNature Communications2041-17232025-08-0116111310.1038/s41467-025-63185-9SpaIM: single-cell spatial transcriptomics imputation via style transferBo Li0Ziyang Tang1Aishwarya Budhkar2Xiang Liu3Tonglin Zhang4Baijian Yang5Jing Su6Qianqian Song7Department of Computer and Information Science, University of Macau, TaipaDepartment of Computer and Information Technology, Purdue UniversityDepartment of Biostatistics and Health Data Science, Indiana University School of MedicineDepartment of Biostatistics and Health Data Science, Indiana University School of MedicineDepartment of Statistics, Purdue UniversityDepartment of Computer and Information Technology, Purdue UniversityDepartment of Biostatistics and Health Data Science, Indiana University School of MedicineDepartment of Health Outcomes and Biomedical Informatics, College of Medicine, University of FloridaAbstract Spatial transcriptomics (ST) technologies have transformed our understanding of cellular organization but are limited by sparse signals and restricted gene coverage. To address these challenges, we introduce SpaIM, a style transfer learning model that leverages single-cell RNA sequencing (scRNA-seq) data to predict unmeasured gene expressions in ST profiles. By disentangling shared content and modality-specific styles, SpaIM effectively integrates scRNA-seq’s rich gene expression with the spatial context of ST. Evaluated across 53 datasets spanning sequencing- and imaging-based spatial technologies in various tissue types, SpaIM consistently outperforms 12 state-of-the-art methods in improving gene coverage and expression accuracy. Furthermore, SpaIM enhances downstream analyses, including ligand-receptor interaction inference, spatial domain characterization, and differential gene expression analysis. Released as open-source software, SpaIM expands accessibility and utility in ST research. Overall, SpaIM represents a robust and generalizable framework for enriching ST data with single-cell information, enabling deeper insights into tissue architecture and cellular function.https://doi.org/10.1038/s41467-025-63185-9
spellingShingle Bo Li
Ziyang Tang
Aishwarya Budhkar
Xiang Liu
Tonglin Zhang
Baijian Yang
Jing Su
Qianqian Song
SpaIM: single-cell spatial transcriptomics imputation via style transfer
Nature Communications
title SpaIM: single-cell spatial transcriptomics imputation via style transfer
title_full SpaIM: single-cell spatial transcriptomics imputation via style transfer
title_fullStr SpaIM: single-cell spatial transcriptomics imputation via style transfer
title_full_unstemmed SpaIM: single-cell spatial transcriptomics imputation via style transfer
title_short SpaIM: single-cell spatial transcriptomics imputation via style transfer
title_sort spaim single cell spatial transcriptomics imputation via style transfer
url https://doi.org/10.1038/s41467-025-63185-9
work_keys_str_mv AT boli spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT ziyangtang spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT aishwaryabudhkar spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT xiangliu spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT tonglinzhang spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT baijianyang spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT jingsu spaimsinglecellspatialtranscriptomicsimputationviastyletransfer
AT qianqiansong spaimsinglecellspatialtranscriptomicsimputationviastyletransfer