EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SpaVGN: A hybrid deep learning framework for high-resolution spatial transcriptomics data reconstruction and spatial domain identification

Haiyan Wang, Yanping Zhang, Yangyang Zhang, Xuening Zhao, Zijia Bai, Xuejing Ma, Chunguang Zhao iD

PLOS One · 14.08.2025 · 20 (8) · e0329122

DOI: 10.1371/journal.pone.0329122 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 14.08.2025. Verfasst von Haiyan Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Physics, Hebei University of Engineering, Handan, China. Hauptthemen laut MEDLINE: Deep Learning, Transcriptome, Gene Expression Profiling. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovative Ability Training Program for Graduate Students in Hebei Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Spatial transcriptomics has revolutionized the analysis of gene expression while preserving tissue spatial information, which provides novel insights into the cellular composition and function of complex biological tissues. However, current technologies are constrained by limited resolution and data sparsity, compromising the accuracy of downstream analyses. To address these challenges, we developed SpaVGN, a deep learning framework integrating convolutional neural networks, vision transformer, and graph neural networks for high-fidelity gene expression imputation and spatial domain identification. By combining local feature extraction, global attention mechanisms, and spatial graph-based modeling, SpaVGN effectively reconstructs missing transcriptomic data while preserving spatial tissue architecture. Evaluated on melanoma and sagittal posterior mouse brain datasets, SpaVGN outperformed existing methods in gene expression prediction, achieving Pearson correlation coefficients of 0.609 (melanoma) and 0.682 (mouse brain). It clearly delineated tumor regions and lymphoid niches in melanoma tissue, achieving fine-grained resolution of hippocampal subfields, including Cornu Ammonis and Dentate Gyrus, with a Silhouette Score of 0.43 and a Davies-Bouldin Index of 0.86. Validation through UMAP dimensionality reduction and PAGA network analysis demonstrated that SpaVGN significantly mitigates the negative impact of data sparsity in spatial transcriptomics, improving data completeness and spatial continuity. This study presents an innovative solution that enhances the resolution of spatial transcriptomics data, offering cross-tissue applicability and providing a valuable tool for research in biological development, disease, and tumor heterogeneity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningTranscriptomeGene Expression ProfilingAnimalsMiceMelanomaBrainNeural Networks, ComputerHumans

Autor:innen und Institutionen

  • School of Mathematics and Physics, Hebei University of Engineering, Handan, China.
  • School of Mathematics and Physics, Handan University, Handan, China.

ORCID-Profile: Chunguang Zhao

Förderung

  • Innovative Ability Training Program for Graduate Students in Hebei Province

Zitieren

APA

Wang, H., Zhang, Y., Zhang, Y., Zhao, X., Bai, Z., Ma, X., & Zhao, C. (2025). SpaVGN: A hybrid deep learning framework for high-resolution spatial transcriptomics data reconstruction and spatial domain identification. PLOS One, 20(8), e0329122. https://doi.org/10.1371/journal.pone.0329122

Vancouver

Wang H, Zhang Y, Zhang Y, Zhao X, Bai Z, Ma X, et al. SpaVGN: A hybrid deep learning framework for high-resolution spatial transcriptomics data reconstruction and spatial domain identification. PLOS One. 2025;20(8):e0329122. doi:10.1371/journal.pone.0329122

BibTeX
@article{Wang2025,
  author = {Wang, Haiyan and Zhang, Yanping and Zhang, Yangyang and Zhao, Xuening and Bai, Zijia and Ma, Xuejing and Zhao, Chunguang},
  title = {SpaVGN: A hybrid deep learning framework for high-resolution spatial transcriptomics data reconstruction and spatial domain identification},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {8},
  pages = {e0329122},
  doi = {10.1371/journal.pone.0329122}
}
Kostenlos starten →