TY - JOUR
T1 - GNTD
T2 - reconstructing spatial transcriptomes with graph-guided neural tensor decomposition informed by spatial and functional relations
AU - Song, Tianci
AU - Broadbent, Charles
AU - Kuang, Rui
N1 - Publisher Copyright:
© 2023, The Author(s).
PY - 2023/12
Y1 - 2023/12
N2 - Spatially-resolved RNA profiling has now been widely used to understand cells’ structural organizations and functional roles in tissues, yet it is challenging to reconstruct the whole spatial transcriptomes due to various inherent technical limitations in tissue section preparation and RNA capture and fixation in the application of the spatial RNA profiling technologies. Here, we introduce a graph-guided neural tensor decomposition (GNTD) model for reconstructing whole spatial transcriptomes in tissues. GNTD employs a hierarchical tensor structure and formulation to explicitly model the high-order spatial gene expression data with a hierarchical nonlinear decomposition in a three-layer neural network, enhanced by spatial relations among the capture spots and gene functional relations for accurate reconstruction from highly sparse spatial profiling data. Extensive experiments on 22 Visium spatial transcriptomics datasets and 3 high-resolution Stereo-seq datasets as well as simulation data demonstrate that GNTD consistently improves the imputation accuracy in cross-validations driven by nonlinear tensor decomposition and incorporation of spatial and functional information, and confirm that the imputed spatial transcriptomes provide a more complete gene expression landscape for downstream analyses of cell/spot clustering for tissue segmentation, and spatial gene expression clustering and visualizations.
AB - Spatially-resolved RNA profiling has now been widely used to understand cells’ structural organizations and functional roles in tissues, yet it is challenging to reconstruct the whole spatial transcriptomes due to various inherent technical limitations in tissue section preparation and RNA capture and fixation in the application of the spatial RNA profiling technologies. Here, we introduce a graph-guided neural tensor decomposition (GNTD) model for reconstructing whole spatial transcriptomes in tissues. GNTD employs a hierarchical tensor structure and formulation to explicitly model the high-order spatial gene expression data with a hierarchical nonlinear decomposition in a three-layer neural network, enhanced by spatial relations among the capture spots and gene functional relations for accurate reconstruction from highly sparse spatial profiling data. Extensive experiments on 22 Visium spatial transcriptomics datasets and 3 high-resolution Stereo-seq datasets as well as simulation data demonstrate that GNTD consistently improves the imputation accuracy in cross-validations driven by nonlinear tensor decomposition and incorporation of spatial and functional information, and confirm that the imputed spatial transcriptomes provide a more complete gene expression landscape for downstream analyses of cell/spot clustering for tissue segmentation, and spatial gene expression clustering and visualizations.
UR - http://www.scopus.com/inward/record.url?scp=85179646663&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85179646663&partnerID=8YFLogxK
U2 - 10.1038/s41467-023-44017-0
DO - 10.1038/s41467-023-44017-0
M3 - Article
C2 - 38092776
AN - SCOPUS:85179646663
SN - 2041-1723
VL - 14
JO - Nature communications
JF - Nature communications
IS - 1
M1 - 8276
ER -