Fig. 3

Analyzing the mouse cerebellum Slide-seqV2 data. A Spatial distribution of all major cell types in the Slide-seqV2 data. Cells are colored by cell types shown in the legend, where the cell type information were predicted using RCTD. B Quantile–quantile plot of the observed −log₁₀(P) from different methods against the expected −log₁₀(P) under the null condition in the permuted Slide-seqV2 data. P values were combined across ten permutation replicates. Permutations were run with SPARK-X with (tan) or without (pale green) adjusting for cell types. C Venn diagram shows the overlap in SE genes identified by SPARK-X with (tan) or without (pale green) adjusting for cell types. D Spatial distribution of predicted cell types (left two panels; blue represents cell type) and spatial expression pattern of four representative SE genes (middle and right panels; green represents high expression while antique-white represents low expression) in Slide-seqV2 data. P values from SPARK-X with (left side of the arrow) or without (right side of the arrow) adjusting for cell types are shown inside parentheses