Extended Data Fig. 7: Molecular differences of malignant epithelial cells between EOPC and LOPC in the validation. | Nature Aging

Extended Data Fig. 7: Molecular differences of malignant epithelial cells between EOPC and LOPC in the validation.

From: Single-cell and spatial RNA sequencing identify divergent microenvironments and progression signatures in early- versus late-onset prostate cancer

Extended Data Fig. 7

(a) Boxplot comparing the AUCell scores of 6 MPs between EOPC (n = 10,249) and LOPC (n = 10,491) malignant epithelia (****PMP2 = 2.1 × 10−12, ****POthers < 2.2 × 10−16). (b) Correlation plot between hypoxia GSVA scores and AR-MP AUCell scores in malignant epithelia. (c) Boxplot of the difference in the ErbB pathways GSVA scores between AR-MP_High (n = 11,062) and AR-MP_Low (n = 9,606) malignant epithelia (****POthers < 2.2 × 10−16). (d) Boxplot of the difference in the lipid metabolism pathways GSVA scores between AR-MP_High and AR-MP_Low malignant epithelia (****POthers < 2.2 × 10−16). (e) Clustering plots of malignant epithelia based on the first 2 components derived from Monocle analysis, colored by pseudo-time (Upper) and AR-MP activation status (Below). (g) Boxplot of the difference in the mRNA stemness indices between AR-MP_High and AR-MP_Low malignant epithelia (****P < 2.2 × 10−16). (g) Boxplot of the difference in the EMT GSVA scores between AR-MP_High and AR-MP_Low malignant epithelia (****P < 2.2 × 10−16). (h) Spatial illustration of AR-MP scores on the Re-EOPC4-ST and Rep-LOPC1-ST slides. The box-and-whisker plots of A, C, D, F and G showed the minimum, 25th percentile, median, 75th percentile and maximum, and data were analysed by two-sided Wilcoxon tests. P-values were not adjusted for multiple comparisons.

Back to article page