Intron-based inference of transcription factor activity allows accurate decoding of dynamic cellular processes

ABSTRACT: Activities of transcription factors (TFs) are temporally modulated to regulate dynamic cellular processes, including development, homeostasis, and disease. Understanding how these processes are regulated thus requires accurate analysis of TF activities. To analyze TF activities, existing methods often leverage the expression levels of the TF’s target genes using transcriptome data. However, because these methods typically use exon-based target expression levels, the estimated TF activities reflect multiple sources of regulation on target genes besides TF activity-dependent regulation, and thus may not faithfully recapitulate actual TF activities. To address this, we devised a TF activity measure using intron-based target expression levels, and implemented it to decode the temporal control of TF activities during dynamic biological processes. Using simulations and public datasets, we first showed that, compared to exon-based estimates, intron-based TF activities display higher correlations with input TF activities, TF nuclear localization levels, and TF chromatin occupancies, suggesting that intron-based measure can better recapitulate instantaneous TF activities. By analyzing public transcriptome data on circadian rhythm and by collecting time-series transcriptome data of T cell response, we found that intron-based TF activities provide an improved characterization of the circadian phasing for cycling TFs in circadian rhythm, and uncover two temporally opposing modules of TFs during T cell’s response to chemical mimics of antigens. Collectively, we anticipate that the proposed approach should be broadly applicable for decoding TF-mediated control of dynamic processes.

ORGANISM(S): Homo sapiens

PROVIDER: GSE178827 | GEO | 2021/12/23

REPOSITORIES: GEO