Extended Data Fig. 3: Protein identification and gene expression in HBV-related early-stage HCC.
From: Proteomics identifies new therapeutic targets of early-stage hepatocellular carcinoma
a, Box plots of the proteins identified in tumour (red, n = 98) and paired non-tumour (green, n = 98) samples. b, Box plots of the protein identifications in AFPlow (grey, n = 51), AFPhigh (red, n = 34), MVI− (grey, n = 51) and MVI+ (red, n = 22) patients from the tumour (red, n = 98) or non-tumour (green, n = 98) samples. To eliminate interaction effects, the patients that are both AFPlow and MVI− are simultaneously considered to be AFPlow or MVI− patients in this analysis. c, Overview of the number of expressed genes in HBV-related early-stage HCC. Similar to the protein identification, the number of genes expressed in the tumour samples is higher than that in the paired non-tumour samples. The MVI+ or AFPhigh patients also have a higher level of gene expression than the MVI− or AFPlow patients in the tumour samples, but there are no statistical differences between the MVI+ or AFPhigh patients and the MVI− or AFPlow patients in the non-tumour samples. The samples are arranged in the same order as in Fig. 1a. Dashed lines denote paired samples from the same cases of HCC. Dashed curves are fitted by lasso regression. The shading that underlies the lasso curves represents the 95% confidence intervals of those lines. d, Box plots of the number of expressed genes in the tumour (red, n = 31) and paired non-tumour (red, n = 31) samples. e, Box plots of the expressed genes in AFPlow (grey, n = 12), AFPhigh (red, n = 14), MVI− (grey, n = 12) and MVI+ (red, n = 9) patients from the tumour (red, n = 31) or non-tumour (green, n = 31) samples. To eliminate interaction effects, patients that are both AFPlow and MVI− are simultaneously considered to be AFPlow or MVI− patients in this analysis. In the box plots, the middle bar represents the median, and the box represents the interquartile range; bars extend to 1.5× the interquartile range. P values are calculated by the two-sided Wilcoxon rank-sum test and are shown on the top of the boxes. NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001.
