Fig. 5: BIT can identify transcriptional regulators from cell-type-specific accessible regions.
From: BIT: Bayesian Identification of Transcriptional regulators from epigenomics-based query region sets
a UMAP projection of human Peripheral blood mononuclear cells (PBMCs) and primary liver cancer cells. b Top 20 TRs identified by BIT using PBMC B cell-specific and liver malignant cell-specific accessible regions, verified by existing literature and Human Protein Atlas (HPA). Posterior estimates of BIT scores are reported with error bars indicating upper and lower bounds of 95% credible intervals. c UMAP visualization of gene activity for PAX5, EBF1, and POU2F1 in PBMCs and HNF4A, FOXA2, and HNF4G in liver cancer samples, each identified by BIT with high activity in B or malignant cells. d Venn plots of top 20 TRs identified by BIT, ArchR, SnapATAC2, and scBasset from PBMC B cells and liver malignant cells. e Gene ontology enrichment results of top 20 TRs identified by BIT, ArchR, SnapATAC2, and scBasset from PBMC B cells. The analysis uses one-sided Fisher exact tests. GO terms are ranked by Benjamini–Hochberg adjusted p values. f Kaplan–Meier Plots of HLF, HNF4A, NFIA, PHOX2B (potential tumor suppressors or differentiation regulators), and CDX2, ETV5, HNF1A, NR1H3 (potential oncogenes) in liver cancer samples. g Numbers of BIT, ArchR, SnapATAC2, and scBasset identified top 10 to 50 TRs from liver malignant cells with minimum Chronos score ≤−0.4. Source data are provided as a Source Data file.
