Fig. 5: MERFISH mouse brain data analysis.

a Data snapshot. Left shows the DAPI image, middle shows expression of four gene sets (EX, blue; IN, green; Astr, red; Olig, orange; Table S12), and right zooms into a subregion showing the same gene sets with cell centroids (crosses) and segmentation boundaries across five z-stacks. b Estimated spatial expression. Upper panel shows gene numbers and proportions, middle shows expression intensities, and lower shows pattern scores. c Example genes and cells for the four pattern clusters. Upper panel lists gene name, P value (1.97e-14, 3.67e-5, 1.23e-4, 2.05e-12), and expression intensities on density maps. Lower panel shows gene expression in five selected cells, overlaid with cell boundaries and aligned nuclear centers (crosses). d Bar plot shows the average sn/sc RNA ratio across genes in clusters 1–2 (red), 3–4 (green; P value = 7e-26), and non-cluster 1–2 genes (i.e., clusters 3–4 plus the nonsignificant genes; gray; P value = 1e-13). Genes enriched close to the nuclear center (clusters 1–2) exhibit higher snRNA levels. e Bar plot displays average gene length, measured by four metrics (x-axis), in pattern clusters 1–2, 3–4, and non-cluster 1–2 genes. Genes enriched close to the nuclear center (clusters 1–2) exhibit longer gene lengths. f Bar plot displays proportions of transcription factors (TFs) for genes in pattern clusters 1–3 (orange), 4 (blue; P value = 6e-5), and non-cluster 4 genes (i.e., clusters 1–3 plus the nonsignificant genes; gray; P value = 1e-3). Genes enriched close to the cell boundary (cluster 4) contain a lower proportion of TFs. g, h Stem plots show the −log10 P values of the top 10 enriched gene sets in GSEA analysis for genes in pattern cluster 3 and 4, respectively. Gene sets enriched with cluster 3 or 4 genes are related to dendrites and synaptic transmission, and signaling. Statistical significance for pair-wise comparisons (*<0.05; **<0.01; ***<0.001; without adjustments for multiple comparisons) is based on two-sided Mann–Whitney U test (d, e) or Fisher’s exact test (f), sample size n = 2923 genes (d, e), data are presented as mean values ± the interquartile range (25th–75th percentile, d, e). Source data are provided as a Source data file.