SIVA: diagonal integration of spatial multi-omics data via spatially informed variational autoencoders and anchor guidance
Bioinformatics · 01.07.2026 · 42 (Supplement_1)
DOI: 10.1093/bioinformatics/btag247 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 07.07.2026. Verfasst von Peng Jiang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence, School of Computer Science, Wuhan University, Wuhan, 430072, China. Hauptthemen laut MEDLINE: Multiomics, Software, Computational Biology. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Lei Jun Foundation Limited, National Natural Science Foundation of China.
Abstract (Originalsprache)
Motivation Understanding cellular states and regulatory programs requires integrative analysis of multiple omics layers. Although recent spatial sequencing technologies allow molecular profiling of cells within their tissue context, paired spatial multi-omics assays are still limited by technical complexity and cost. This creates a pressing need for diagonal integration methods that enable joint analysis of unpaired spatial omics datasets. Results We propose SIVA, a deep generative framework based on Spatially-Informed Variational Autoencoders with Anchor Guidance, for diagonal integration of spatial multi-modal data. SIVA employs modality-specific variational autoencoders (VAEs) with a hybrid latent embedding that integrates Gaussian process and standard Gaussian priors, enabling joint modeling of spatially structured variation and dominant underlying data distributions across modalities. To facilitate cross-modal alignment in the absence of one-to-one cell correspondence, SIVA adopts a dual integration strategy combining global distribution alignment via Maximum Mean Discrepancy and local correspondence guidance using mutual nearest neighbor anchors. Extensive experiments across multiple cross-slice integration scenarios demonstrate that SIVA achieves robust and accurate integration of unpaired spatial omics datasets, consistently outperforming existing methods. Availability and implementation The source codes are available at https://github.com/PelenJiang/SIVA
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Artificial Intelligence, School of Computer Science, Wuhan University, Wuhan, 430072, China.
Förderung
- Lei Jun Foundation Limited
- National Natural Science Foundation of China
Zitieren
APA
Jiang, P., Chen, S., Wu, X., Liu, J., & Tian, T. (2026). SIVA: diagonal integration of spatial multi-omics data via spatially informed variational autoencoders and anchor guidance. Bioinformatics, 42(Supplement_1). https://doi.org/10.1093/bioinformatics/btag247
Vancouver
Jiang P, Chen S, Wu X, Liu J, Tian T. SIVA: diagonal integration of spatial multi-omics data via spatially informed variational autoencoders and anchor guidance. Bioinformatics. 2026;42(Supplement_1). doi:10.1093/bioinformatics/btag247
BibTeX
@article{Jiang2026,
author = {Jiang, Peng and Chen, Sishuo and Wu, Xingye and Liu, Juan and Tian, Tian},
title = {SIVA: diagonal integration of spatial multi-omics data via spatially informed variational autoencoders and anchor guidance},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {Supplement_1},
doi = {10.1093/bioinformatics/btag247}
}
EUVIMED