SEEK to Identify Super Enhancer-Expression Correlated Pairs using Single-cell Multi-omic Data

Abstract

Super enhancers (SEs) drive cell identity and disease related genes. However, current methods for studying associations between SE and gene expression are time consuming, costly and with poor scalability. This study formulated a computational approach for screening genome-wide SE-expression associations by analyzing single-cell multi-omic data of transcriptome and H3K27ac histone modification. A pipeline was also constructed for an easy workflow application. Further our application study identified expression correlated SEs (eSEs) in brain and found they mark cell types. Moreover, our analysis provided new insights into the functional role of SEs close to Kcnip4 and Nifb1 in frontal cortex neurons and CGE derived inhibitory neurons, linking to neuron development and neurological diseases. Collectively, this study provides a new tool for studying SE-expression associations and identifying significant expression associated SEs, which pave the way for understanding the regulatory role of SEs in gene expression and related cellular and disease development.