Fig. 2

Concordance, statistical significance, and overlap of SVGs detected by different methods. a Concordance of SVG rankings reported from each SVG detection method. Each panel uses one SVG detection as an anchor and the y-axes are pairwise Spearman’s correlation coefficient for quantifying concordance in ranking of each pair of SVG detection methods. Points in each boxplot represent the result from a dataset. Boxplot centre line, median; box limits, upper and lower quartiles; whiskers, 1.5 times the interquartile range. b Statistical significance of SVGs detected by each method. SVGs are partitioned into three categories based on the adjusted p-values reported by each method (i.e., p = 0; 0 < p ≤ 0.05; p > 0.05) and presented as a percentage (y-axis). The datasets are ordered in terms of the decreasing proportion of genes observed in the orange category. The color bars denote various characteristics of the spatial dataset including the tissue type, spatial technology, number of spatial locations, and total number of genes expressed. c A proportional bar plot showing the percentage of unique (# of method = 1) and overlapping (# of method > 1) significant SVGs (adjusted p-value ≤ 0.05) reported by the SVG detection methods for each spatial transcriptomics dataset