Fig 1.
Clinicopathological features of tumour samples in the Korean cohort.
This includes the surgery type (TH/PH), disease recurrence, whether the tumour sample has an adjacent normal sample, the premalignant stage of the tumour-adjacent normal, and the risk factor (HBV, HCV, Alcoholic, None, SLD).
Fig 2.
The workflow to study the progression from normal to precancerous to cancer state in HCC.
Fig 3.
(A) Module-trait correlations of tumour samples. DFS representing disease-free survival is a continuous variable. Surgery (treatment) is a binary variable with partial hepatectomy (PH) represented as 1 and total hepatectomy (TH) as 2. *** indicates p-value < 0.001, ** indicates 0.001 ≤ p-value < 0.01, * indicates 0.01 ≤ p-value < 0.05. (B) KEGG pathway enrichment of tumour modules. For each module, 15 most significant pathways sorted according to adjusted p-value are displayed (bottom to top within each module). *** indicates adjusted p-value < 0.001, ** indicates 0.001 ≤ adjusted p-value < 0.01, * indicates 0.01 ≤ adjusted p-value < 0.05. The number of overlapping genes and the total number of pathway genes are shown to the right of each bar. (C) Survival analysis based on eigengene expression of tumour modules. Samples are classified into high and low-expression groups based on the median of eigengene expression of each module. ‘p’ indicates the p-value of survival analysis.
Fig 4.
Progression from precancerous to cancer state.
Module-trait correlations in (A) TH and (B) PH treatment groups. Tissue type is a binary variable with the precancerous state as 1 and cancer state as 2. *** indicates p-value < 0.001, ** indicates 0.001 ≤ p-value < 0.01, * indicates 0.01 ≤ p-value < 0.05. (C) KEGG pathway enrichment of precancerous to cancer modules. For each module, the 15 most significant pathways sorted according to adjusted p-value are displayed (bottom to top within each module). *** indicates adjusted p-value < 0.001, ** indicates 0.001 ≤ adjusted p-value < 0.01, * indicates 0.01 ≤ adjusted p-value < 0.05. The number of overlapping genes and the total number of pathway genes are shown to the right of each bar.
Fig 5.
Co-expression modules of normal and premalignant conditions.
(A) Module-trait correlations of premalignant samples. The stage represents different premalignant conditions. (B) Eigengene plots for individual premalignant modules showing correlation with premalignant stage. *** indicates p-value < 0.001, ** indicates 0.001 ≤ p-value < 0.01, * indicates 0.01 ≤ p-value < 0.05.
Fig 6.
Eigengene-based survival analysis using tumour-adjacent normal samples in the Chinese cohort.
The eigengene expression of each premalignant module (obtained from the Korean cohort) was calculated using the tumour-adjacent normal samples in the Chinese cohort. Samples are classified into high and low-expression groups based on the median of eigengene expression of each module. ‘p’ indicates the p-value of survival analysis.
Fig 7.
Cell cycle alterations from normal to HCC transition.
(A) Venn diagram showing the overlap of cell cycle genes with modules significantly enriched for cell cycle-related pathways in premalignant and malignant samples (PH and TH groups). (B) Network of GO biological processes of cell cycle genes in premalignant module N4. (C) Network of GO biological processes of cell cycle genes in precancerous-cancer module TH3.
Fig 8.
Network of GO biological processes of cell cycle genes in precancerous-cancer module PH2.