NLM DIR Seminar Schedule

Abstract:

Transcription factors (TFs) play pivotal roles in gene regulation by binding to regulatory DNA elements like enhancers, dictating the spatial and temporal expression of their target genes. Precise identification of transcription factor binding sites (TFBSs) is crucial for linking genetic variants to complex human traits or diseases. Traditional computational methods often over-annotate enhancers as binding sites, leading to a high rate of false positives. In this study, we apply a deep learning approach for accurately identifying TFBSs within liver enhancers, covering less than 10% of enhancer regions. Our model effectively captures TFBSs of key activator TFs specific to the liver, but ATAC-seq footprinting does not. Notably, we observe optimal clustering of TFBSs associated with activator TFs based on motif similarity. In contrast, TFBSs linked to repressor TFs do not exhibit such clustering, suggesting a diverse repertoire of TFs acting as enhancer repressors. These findings shed light on the nuanced regulatory landscape within enhancers and underscore the importance of advanced computational techniques in deciphering transcriptional regulatory networks.