Extended Data Fig. 9: Delineating the polarity and biological patterning of X-Y gametolog functional divergence.

a. PCA of asymmetric coupling values across all pairs–tissues (represented by each point, see legend). b. tSNE of asymmetric coupling values across all pairs–tissues (represented by each point, see legend). c. UMAP of asymmetric coupling values across tall pairs–tissues (represented by each point, see legend). d. PCA of asymmetric coupling values across all pairs–tissues (represented by each point, see legend). Similar tissues have been collapsed (Methods). e. tSNE of asymmetric coupling values across all pairs–tissues (represented by each point, see legend). Similar tissues have been collapsed (Methods). f. UMAP of asymmetric coupling values across all pairs–tissues (represented by each point, see legend). Similar tissues have been collapsed (Methods). g. Mean asymmetric coupling values with N = 15 gametolog pairs (excluding AMELX/Y and RPS4X/Y2, which are not expressed) for genes in each cluster (averaged across tissues). N = 8 clusters were derived from N = 11,498 genes whose asymmetric coupling values were predicted by gametolog pair only (see Fig. 4g) (clusters with < 20 genes were removed). Clusters and pairs are ordered according to hierarchical clustering. Asterisks indicate p_adj < 0.05 (versus null, two-sided; Methods). h. Extended Data Fig. 9g, but with mean sCFD_XM–YM values shown for each tissue. i. Extended Data Fig. 9g, but with pairs ranked within each cluster according to their mean sCFD_XM–YM values. j. Top enriched biological processes and associated −log10(p-values) for each cluster. Top 2 categories are shown for each cluster. k. Top enriched cellular compartments and associated −log10(p-values) for each cluster. Top 2 categories are shown for each cluster. l. Similar to Extended Data Fig. 9g, but for genes whose expression is predicted by tissue only (see Supplementary Table 13, Column L). Mean asymmetric coupling values in each tissue for genes in each cluster. N = 3 clusters were derived from N = 40 genes whose asymmetric coupling values were predicted by tissue only (Fig. 4g). Clusters and pairs are ordered according to hierarchical clustering. Asterisks indicate p_adj < 0.05 (versus null, two-sided; Methods). m. Extended Data Fig. 9l, but with mean sCFD_XM–YM values shown for each pair. n. Extended Data Fig. 9l, but with tissues ranked within each cluster according to their mean sCFD_XM–YM values. o. Extended Data Fig. 9j, but for genes whose expression is predicted by tissue only (see Supplementary Table 13, Column L). Top enriched biological processes and associated −log10(p-values) for each cluster. Top 4 categories are shown for each cluster. p. Extended Data Fig. 9k, but for genes whose expression is predicted by tissue only (see Supplementary Table 13, Column L). Top enriched cellular compartments and associated −log10(p-values) for each cluster. Top 4 categories are shown for each cluster. q. Similar to Extended Data Fig. 9g, but for genes whose expression is predicted by both pair and tissue (see Supplementary Table 13, Column M). Mean asymmetric coupling values for each pair (top) or tissue (bottom) for genes in each cluster. N = 5 clusters were derived from N = 1,141 genes whose asymmetric coupling values were predicted by pair and tissue (Fig. 4g). Clusters and pairs are ordered according to hierarchical clustering. Asterisks indicate p_adj < 0.05 (versus null, two-sided; Methods). r. Extended Data Fig. 9q, but with mean sCFD_XM–YM values shown for each tissue (top) or pair (bottom). s. Similar to Extended Data Fig. 9j, but for genes whose expression is predicted by pair and tissue (see Supplementary Table 13, Column M). Top enriched biological processes and associated −log10(p-values) for each cluster. Top 2 categories are shown for each cluster. t. Similar to Extended Data Fig. 9k, but for genes whose expression is predicted by pair and tissue (see Supplementary Table 13, Column M). Top enriched cellular compartments and associated −log10(p-values) for each cluster. Top 2 categories are shown for each cluster.