ChEA-KG and ChEA-KG-TS: a network-based transcription factor enrichment analysis tool with an accompanying time-series workflow.

Abstract

Transcription factor (TF) modules interact to regulate key biological processes and cell-state transitions in normal physiology and disease. Understanding these modules and how they evolve over time can be accomplished by constructing gene regulatory networks (GRNs). To identify context-specific TF subnetworks, we developed ChEA-KG, which generates enriched TF regulatory subnetworks for input gene sets. ChEA-KG is based on a GRN connecting 1559 human TFs via 131 181 signed and directed edges inferred from diverse published ChIP-seq (chromatin immunoprecipitation followed by sequencing) and mRNA (messenger RNA)-sequencing experiments. We demonstrate ChEA-KG's utility by applying it to uncover master regulators of aging, mechanisms of action (MoA) for drug classes, pan-cancer subtypes, and cell types from across 14 major human tissues. Next, we extend ChEA-KG to develop the webserver application ChEA-KG Time Series (ChEA-KG-TS), which identifies TF modules from time-series mRNA-sequencing datasets. Results from this workflow are automatically summarized as reports that include enrichment analysis, regulatory subnetworks, and UMAP projections of enriched TFs. We use ChEA-KG-TS to explain transient responses in two use cases. ChEA-KG and ChEA-KG-TS are available from https://chea-kg.maayanlab.cloud/ and https://chea-kg-ts.maayanlab.cloud/.