Fig 1.
Cluster of cluster analysis of HCC samples.
The a) heatmap of subclustersfronplatformed were analyzed and subgroups were identified. We found that the b) survival rate of subgroups were significantly different, and the c) mutation of some genes, especially TP53 and CTNNB1.
Table 1.
Association between clinical observation and subgroups.
Fig 2.
Multiple alterations were observed in S1.
UGT2B17 deletion is significantly severe in S1 than S2-S4. Subcluster mRNA4 is enriched in S1, and b) mRNA4 genes were enriched in GO terms involved in lipid metabolism. On methylation level, c) DCDC gene promoter region were hypomethylated, as indicted by the black rectangle. Onco-proteins were d) highly expressed in S1, while e) tumor suppressor proteins were lowly expressed in S1.
Fig 3.
Telomere hypo-methylation was a characteristic of S2 and S3.
a) The distribution of methylation clusters. Among them, Methyl1/2/6/7 were hypermethyalted, the others were hypomethylated (shown in Fig 1A). Prognostic marker b) CRP/CYP2E1, and c) miRNA mir-199a-1/mir-199a-2/mir-199b were significantly different in S2 and S3. Consistent with the hypomethylation of telomere, enzymes involved in d) telomere elongation and methylation were down regulated.
Fig 4.
a) Promoter hypermethylation of MARCH11 is detected in S4. On transcriptomic level, prognostic marker b) mir-203 and c) AXIN2 were highly/lowly expressed in S4, respectively.
Fig 5.
Expression level of mRNA10/12/15 and GO term of the included genes.
The expression heat map of mRNA10/12/15 cluster (a-c, left panel) and the rasult of GO analysis of the related genes (a-c, right panel).
Table 2.
Summary of molecular characters of S1-S4.
Table 3.
The AUC of character features for each subgroup.