Fig. 4: Male–female differences in cell types.

a–d t-SNE representation of sn transcript expression (a, males; b, females) and of sn chromatin accessibility (c, males; d, females) in merged same-sex samples, with labeling by individual animal. Individual animals are color-coded as indicated. Each cluster is identified by a letter code. The code for each cell type is presented at the bottom of Fig.2. In panel d, magnifications of the proliferative somatotrope and proliferative lactotrope clusters are shown as inserts. Refer to Fig.2a,b (males) and Supplementary Fig. 2a,b (females) for the merged analyses. e, f Cell type proportions per animal identified from the snRNAseq (a, b) and snATACseq (c, d) datasets show individual differences that are concordant (see also Supplementary Table 2). Male–female differences are significant for lactotropes and somatotropes (p < 0.0001) in both assays. N = 6 independent animals were examined (3 males, 3 females), with the following number of nuclei per assay and per animal: (i) for snRNAseq: male #1, 7,150; male #2, 4,562; male #3, 6,146; female #1, 6,717; female #2, 5,282; female #3, 4,186; (ii) for snATACseq: male #1, 2,782; male #2, 880; male #3, 1,095; female #1, 4,847; female #2, 9,976; female #3, 11,832. Data are presented per individual animal, with a single column for each. For the assessment of sex difference in cell type proportions, we used a two-way analysis of variance (ANOVA) followed by Bonferroni multiple comparisons post-hoc test, with n = 3 biological replicates per cell type. g Cell type proportions identified in snMethyl data from pooled male pituitaries. A pool of 30 animals was used with 2,756 nuclei analyzed. h 3D correlation plot between gene expression, chromatin accessibility, and snMethyl for the identification of cell type proportions in male samples.