Loading
Spatially aware clustering groups the spots of a tissue section into domains using both what they express and where they sit, so the domains are contiguous regions rather than scattered spots.
SpaGCN, from Hu and colleagues, uses a graph convolutional network over spots linked by spatial proximity and histology to identify spatial domains and their marker genes. Simpler versions average each spot with its k nearest neighbours (Wikipedia on k-NN) before principal components and k-means clustering (Wikipedia), which reduces the sparsity of spatial counts and yields coherent regions at the cost of blurring boundaries. Moran's I then tests whether the resulting labels or risk scores are spatially structured.
Showing the technology this term belongs to: Spatial transcriptomics.
Shares Spatial autocorrelation (Moran's I), Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Spatial transcriptomics, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Principal component analysis (PCA) as a feature compressor, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Principal component analysis (PCA) as a feature compressor, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Principal component analysis (PCA) as a feature compressor, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Principal component analysis (PCA) as a feature compressor, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Principal component analysis (PCA) as a feature compressor, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Spatial transcriptomics platforms (Visium HD, Xenium, MERFISH, CosMx, CODEX), Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.