Abstract:
Enhancers and their non-coding counterparts, enhancer RNAs (eRNAs), are crucial regulators of gene expression, directing cellular identity, developmental programs, and disease states. The functional implications of enhancer perturbation, whether performed via knockout or knockdown, are still incompletely characterized because not all perturbations produced measurable changes in gene expression. To address this, we developed ePerturbDB (http://reggen.iiitd.edu.in:1210/), a manually curated, open-access repository of 83,743 enhancer and eRNA perturbation records collected from experimental literature in diverse biological contexts. ePerturbDB provides users with the ability to compare user query genomic loci with available validated perturbations and interrogate the associated genes and ontology terms to infer possible regulatory implications, thereby enabling both functional annotation and translational therapies. Simultaneously, we implemented a machine learning methodology predicting gene-pathway-enhancer association based on enhancer activity profiles to find enhancers that regulate genes within defined biological pathways. These predictions provide insight into enhancer biology and collectively serve as a tool to explore the enhancer-mediated gene regulation and support the foundation of designing functional studies in the contexts of genomics and precision medicine.
