Figure 3 | Scientific Reports

Figure 3

From: Integrated genomic/epigenomic analysis stratifies subtypes of clear cell ovarian carcinoma, highlighting their cellular origin

Figure 3

Methylation-based clustering of ovarian carcinomas (CCOC, HGSOC, MOC, and EMOC) and normal tissues. (A) Comparison of non-negative matrix factorization consensus clustering of DNA methylation between ovarian cancers and normal tissues (FTE, PPM, OSE, and EME) using significant differentially methylated probes. Our analysis includes CCOC (n = 92) and HGSOC (n = 83) samples alongside publicly available samples (11 CCOC, 65 HGSOC, 8 mucinous carcinomas, and 11 endometrioid carcinomas). (B) Principal component analysis of cancers (green, red, and blue dots are cancers in the FTE-like, PPM-like, and OSE/EME-like clusters, respectively) and normal tissues (shown in F, P, O, and E). (C) Methylation heat map sorted through unsupervised hierarchal clustering in PPM-like CCOC. The mutational status of ARID1A and PIK3CA is shown for each sample. (D) The list of enriched motifs detected in hyper-methylated enhancer regions in the SWI/SNF alteration cluster.

Back to article page