Fig. 1

Identification of potential mRNA biomarkers for iCCA patients. A Venn plot showing iCCA/HCC-specific mRNA biomarkers among differentially expressed genes between nontumor liver tissue and iCCA or HCC tissue samples from TCGA, GEO or ArrayExpress datasets, respectively. TCGA-HCC dataset including 50 nontumor liver tissue samples and 371 HCC tissue samples, TCGA-iCCA dataset including 8 nontumor liver tissue samples and 32 iCCA tissue samples, E-MTAB-6389 dataset including 31 nontumor liver tissue samples and 78 iCCA tissue samples, and GSE26566 dataset including 65 nontumor liver tissue samples and 104 iCCA tissue samples. Gallbladder or extrahepatic carcinoma samples (n = 4) in TCGA cholangiocarcinoma dataset were excluded from our analysis. B Hierarchical clustering dendrogram showing gene expression modules in E-MTAB-6389 dataset including 1669 iCCA-specific mRNA biomarkers. C Heatmap to show the correlation between gene expression modules and clinical features of patients with iCCA enrolled in E-MTAB-6389 cohort. Value represents coefficient of Pearson correlation test, and the number in parentheses denotes statistical significance. D Heatmap to show the expression patterns of iCCA-specific mRNA biomarkers from brown module (MEbrown) in iCCA and nontumor liver tissue samples from E-MTAB-6389 dataset. E GO enrichment analysis showing biological processes significantly enriched by iCCA-specific mRNA biomarkers from brown module. F Volcano plot to show the expression fold change of 76 differentially expressed genes from brown module. iCCA, intrahepatic cholangiocarcinoma; HCC, hepatocellular carcinoma; GEO, Gene Expression Omnibus; TCGA, The Cancer Genome Atlas