Dependency-aware deep generative models for multitasking analysis of spatial omics data
Document Type
Article
Publication Date
8-1-2024
Abstract
Spatially resolved transcriptomics (SRT) technologies have significantly advanced biomedical research, but their data analysis remains challenging due to the discrete nature of the data and the high levels of noise, compounded by complex spatial dependencies. Here, we propose spaVAE, a dependency-aware, deep generative spatial variational autoencoder model that probabilistically characterizes count data while capturing spatial correlations. spaVAE introduces a hybrid embedding combining a Gaussian process prior with a Gaussian prior to explicitly capture spatial correlations among spots. It then optimizes the parameters of deep neural networks to approximate the distributions underlying the SRT data. With the approximated distributions, spaVAE can contribute to several analytical tasks that are essential for SRT data analysis, including dimensionality reduction, visualization, clustering, batch integration, denoising, differential expression, spatial interpolation, resolution enhancement and identification of spatially variable genes. Moreover, we have extended spaVAE to spaPeakVAE and spaMultiVAE to characterize spatial ATAC-seq (assay for transposase-accessible chromatin using sequencing) data and spatial multi-omics data, respectively.
Identifier
85193988508 (Scopus)
Publication Title
Nature Methods
External Full Text Location
https://doi.org/10.1038/s41592-024-02257-y
e-ISSN
15487105
ISSN
15487091
PubMed ID
38783067
First Page
1501
Last Page
1513
Issue
8
Volume
21
Grant
BK20230781
Fund Ref
Natural Science Foundation of Jiangsu Province
Recommended Citation
Tian, Tian; Zhang, Jie; Lin, Xiang; Wei, Zhi; and Hakonarson, Hakon, "Dependency-aware deep generative models for multitasking analysis of spatial omics data" (2024). Faculty Publications. 261.
https://digitalcommons.njit.edu/fac_pubs/261