Fig. 2: SIMVI reveals intrinsic and spatial variations in MERFISH human cortex and Slide-seqV2 mouse hippocampus. | Nature Communications

Fig. 2: SIMVI reveals intrinsic and spatial variations in MERFISH human cortex and Slide-seqV2 mouse hippocampus.

From: SIMVI disentangles intrinsic and spatial-induced cellular states in spatial omics data

Fig. 2

a Overview of the MERFISH middle temporal gyrus (MTG) data, showing spatial organizations of cell type, layer annotation, and log normalized MYH11 expression for MTG replicates 1-3. EXC excitatory neurons, INH inhibitory neurons, ASC astrocytes, MGC microglial cells, OGC oligodendrocytes, OPC oligodendrocyte progenitor cells, ENDO endothelial cells, MURAL mural cells. L1-L6, Layer 1-6. b Bar plots showing metric scores for the MERFISH MTG and superior temporal gyrus (STG) datasets respectively (See Methods for definitions of the scores). The bar heights represent the average performance across 10 different random seeds, with the error bars showing standard errors. c UMAP visualization of the SIMVI intrinsic variation, colored by cell types and replicate labels. d UMAP visualization of the SIMVI spatial variation, colored by layer annotation, log normalized MYH11 expression, and replicate labels. Rep 1-3, Replicate 1-3. e Overview of the Slide-seqv2 data showing spatial organization of pixels (left) and Allen Mouse Brain Atlas annotation for the region (right). CA Cornu Ammonis, cl cluster, DG Dentate Gyrus, V3 Third Ventricle, MH Medial Habenula, LH Lateral Habenula. f Spatial visualization of clusters obtained from SIMVI, CellCharter and GraphST. g Bar plot showing the metric scores for the Slide-seqV2 mouse hippocampus data (See Methods for definitions of the scores). The bar heights represent the average performance across 10 different random seeds (n = 9 for SIMVI), with the error bars showing standard errors. †: Each individual metric from all experiments is scaled to the range [0,1], then the final score is calculated as the average of these rescaled values. *: The score is obtained by first averaging the individual score values for each experiment, then rescaling the average score across all experiments to the range [0,1]. Source data are provided as a Source Data file.

Back to article page