# IVEA: an integrative variational Bayesian inference method for predicting enhancer–gene regulatory interactions

## TL;DR

IVEA is a new computational method that predicts how enhancers regulate genes using transcription data and chromatin features.

## Contribution

IVEA introduces a variational Bayesian inference framework based on transcriptional bursting to infer enhancer–gene interactions.

## Key findings

- IVEA achieves high accuracy in predicting enhancer–gene regulatory relationships.
- The method provides biologically relevant insights into enhancer–gene interactions.

## Abstract

Enhancers play critical roles in cell-type-specific transcriptional control. Despite the identification of thousands of candidate enhancers, unravelling their regulatory relationships with their target genes remains challenging. Therefore, computational approaches are needed to accurately infer enhancer–gene regulatory relationships.

In this study, we propose a new method, IVEA, that predicts enhancer–gene regulatory interactions by estimating promoter and enhancer activities. Its statistical model is based on the gene regulatory mechanism of transcriptional bursting, which is characterized by burst size and frequency controlled by promoters and enhancers, respectively. Using transcriptional readouts, chromatin accessibility, and chromatin contact data as inputs, promoter and enhancer activities were estimated using variational Bayesian inference, and the contribution of each enhancer–promoter pair to target gene transcription was calculated. Our analysis demonstrates that the proposed method can achieve high prediction accuracy and provide biologically relevant enhancer–gene regulatory interactions.

The IVEA code is available on GitHub at https://github.com/yasumasak/ivea. The publicly available datasets used in this study are described in Supplementary Table S4.