Skip to main content
. 2022 Jun 27;2(6):387–398. doi: 10.1038/s43588-022-00263-8

Extended Data Fig. 1. Application of ActiveSVM to identify region specific marker genes in the mouse brain with spatial transcriptomic data.

Extended Data Fig. 1

The results of classification where cells are labeled according to fields of view (FOV). (a-b) test accuracy with min-complexity strategy, where comparison methods use the same number of cells as ActiveSVM in (a) and use the entire dataset in (b). Fields of view 1-5 correspond to 5 regions of the mouse cortex, additional fields of view are labeled SVZ (subventricular zone) and ChP (choroid plexus). (c) t SNE of cell transcriptomes for all cells (d) number of cells used per iteration (e) Sample of identified genes where each sub-panel shows mean expression across FOV/brain regions for selected gene, a tSNE plot colored by expression of selected gene, a violin plot of single cell gene expression values for selected gene in FOV/brain region, and spatial plots of each field of view where dots represents cells in 2D imaging slice, cells are colored by intensity of selected gene and units are in millimeters.

Source data