Skip to main content
Fig. 6 | Genome Biology

Fig. 6

From: HyperChIP: identification of hypervariable signals across ChIP-seq or ATAC-seq samples

Fig. 6

Applying HyperChIP to a TCGA pan-cancer ATAC-seq data set. a Two-dimensional t-SNE plot of all patients based on their ATAC-seq profiles of tumor tissues. The patients are colored by cancer types (see Additional file 3: Table S3 for the full names of the involved cancer types). b The distribution of the patients belonging to the SC class. c The distribution of the patients belonging to the DIAD class. d The distribution of all ESCA patients, comprising 12 ESSC and 6 ESAD cases. e Heat map showing the TF activity scores of the top 15 binding motifs identified for each class of cancer types. f Plotting the t-statistics of all motifs against their rankings in the identification of DIAD class-specific TFs. g Mapping the TF activity scores of the HNF4A motif to the t-SNE plot. h Mapping the expression levels of the HNF4A gene (calculated from the corresponding RNA-seq samples) to the t-SNE plot. i Box plots showing the expression of HNF4A in a larger TCGA cohort of (7183) patients. The expression data are accessed via the online tool GEPIA, in which ESSC and ESAD patients are both labeled ESCA and cannot be distinguished from each other. The cancer types are sorted by the median expression of HNF4A in tumors. For each cancer type, the significance of differential expression is determined by performing a t-test with a p-value cutoff of 0.01 and a fold change cutoff of 2. TPM, transcripts per million. j Plotting the t-statistics of all motifs against their rankings in the identification of SC class-specific TFs. k Mapping the TF activity scores of the TP63.1 motif to the t-SNE plot. l Mapping the expression levels of the TP63 gene to the t-SNE plot. m Box plots showing the expression of TP63 in the larger TCGA cohort

Back to article page