Back to search

Cornell University

Genome-wide identification of cell-type-specific active enhancers

Abstract

dc:description.abstract

Enhancers play an essential role in regulating gene expression, and their identification is crucial for understanding molecular mechanisms and disease etiology. Previous studies relied on epigenomic marks, such as chromatin accessibility and histone modifications (e.g., H3K27ac, H3K4me1), which may not offer the specificity necessary for precise enhancer detection. Recent advances have revealed that enhancers can transcribe their own RNAs, known as eRNAs, providing a novel approach for identifying active enhancers through RNA sequencing. With a variety of sequencing assays available, selecting one that adequately addresses the low abundance and short half-life of eRNAs is critical for accurately identifying active enhancers genome-wide. This dissertation compared 13 RNA sequencing assays for detecting eRNAs, demonstrating that run-on assays such as GRO/PRO-cap excel in sensitivity and resolution due to their ability to capture nascent transcripts. To better identify transcribed enhancers, the computational tool Peak Identifier for Nascent Transcription Starts (PINTS) was developed, designed specifically for libraries generated by assays profiling transcription initiation sites of the nascent transcriptome. Application of PINTS led to a comprehensive compendium of transcribed enhancers across diverse tissues and cell types, providing an invaluable resource for studying gene regulation. Despite its strengths, GRO/PRO-cap requires substantial input material similar to many other bulk assays, limiting its utility in dissecting cell-type-specific regulatory patterns in tissue samples. To overcome this limitation, the quasi-supervised cross-modality deconvolution framework, Deep-learning-based DEconvolution of Tissue profiles with Accurate Interpretation of Locus-specific Signals (DeepDETAILS), was introduced. This model reconstructs cell-type-specific profiles from bulk sequencing libraries using a reference scATAC-seq library. DeepDETAILS can deconvolve signals captured from various regulatory steps, including transcription initiation, pause-release, and histone modifications, demonstrating its broad applicability. By employing DeepDETAILS, a comprehensive compendium of cell-type-specific regulatory maps was created, encompassing 39 human tissues and 86 distinct cell types. This resource facilitated the identification of a risk variant potentially contributing to the etiology of primary sclerosing cholangitis. Together, this dissertation identified cell-type-specific enhancers genome-wide, advancing our understanding of transcriptional regulation and its implications for cellular function and disease mechanisms.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Computational Biology
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Computational Biology
Grantor
Cornell University
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Yao, Li
Chair dc:contributor.chair
  • Yu, Haiyuan
Committee members dc:contributor.committeemember
  • Lis, John
  • Booth, James

Subjects

dc:subject × 6

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16938340
Dc Identifier Other
ProQuest Submission ID: 14861
ProQuest Publication ID: 31937148

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Yao, Li. Genome-wide identification of cell-type-specific active enhancers. Doctor of Philosophy thesis, Cornell University, 2025. https://hdl.handle.net/1813/117670