UMass Chan scientists annotate largest map yet of human genome’s regulatory switches

A research team led by Zhiping Weng, PhD, and Jill Moore, PhD’18, at UMass Chan Medical School, has nearly tripled the known number of potential regulatory elements in the genome to 2.37 million, creating the most comprehensive map to date of the DNA sequences that control when and where genes are turned on and off in human cells, as published in Nature.

These genetic switches, called cis-regulatory elements (CREs), are found outside the main protein coding sequences of genes, and control the transcription of nearby genes. CREs are critical for specialized genetic and biological functions. Using this registry of candidate CREs (or cCREs), Drs. Moore and Weng and their teams were able to show how large-scale regulatory maps can reveal previously unrecognized classes of CREs and illuminate how noncoding genetic variation contributes to cell type-specific traits.

“Understanding the biological contexts and functions of CREs is essential for deciphering genomic function and its impact on health and disease,” explained Weng, the Li Weibo Chair in Biomedical Research and chair and professor of genomics & computational biology. “This directory of candidate cis-regulatory elements is a powerful resource for researchers studying the role gene regulation plays in the formation of complex traits and diseases.”

The study was carried out as part of the Encyclopedia of DNA Elements project, known as ENCODE. A two-decade international effort, organized into four successive five-year phases, ENCODE works to systematically catalog the functional elements of the human and mouse genomes. Weng has served as a principal investigator in all four phases of ENCODE and has led the consortium’s data analysis center for the last 10 years, helping to coordinate the large-scale integration and public release of its datasets and analysis results.

While only a small fraction of the human genome encodes for proteins, much of the remaining DNA serves in a regulatory capacity.

“The genome contains millions of short regulatory sequences that act as control elements for genes, turning them on or off in specific cell types, developmental stages or environmental conditions,” explained Dr. Moore, assistant professor of genomics & computational biology. “These CREs are central to almost every biological process, and many disease-associated genetic variants fall within these sequences.”

Systematic maps of these regulatory elements, however, remain incomplete. Using a range of biochemical benchmarks identified with biological assays, improved analysis and new external datasets, the latest study has expanded the number of candidate regulatory sites from 900,000 in the human genome and 300,000 in the mouse genome to 2.37 million and 927,000 respectively.   

“More importantly, subsets of these elements are selectively active in different cell and tissue types, including developmental and disease-relevant states, providing a reference atlas of where regulatory DNA elements resides and in which cellular contexts they are active,” explained Moore.

A major advance of this work is the scale of functional characterization. Rather than inferring regulatory activity solely from chromatin structure, the team integrates data from a broad collection of high-throughput assays. In all, 5,712 human experiments and 758 mouse experiments, performed by the ENCODE Consortium, were used to annotate the cCRE registry.

“Taken together, these datasets provide some form of functional measurement for more than 90 percent of human regulatory elements,” said Weng. “This integration reveals not only classical enhancers that increase gene expression, but also a large number of DNA elements that act as silencers to repress genes.”

Researchers also found that some regulatory elements can function as enhancers in one cell type and as silencers in another, depending on the combination of transcription factors present. This context-dependent, dual-function behavior highlights the flexibility of regulatory DNA sequences from cell to cell. 

The expanded registry also serves as a powerful framework for interpreting the function of noncoding DNA sequence variants associated with human traits and diseases. Genome-wide association studies (GWAS) have identified tens of thousands of variants that are associated with common conditions such as heart disease, diabetes and schizophrenia, but most of these variants lie outside protein-coding genes. By overlaying GWAS signals onto a map of potential regulator elements that are active in trait-relevant cell types and then using results from multiple high-throughput assays, researchers can link these regulatory elements to their likely target genes.

Using this strategy on red blood cells, Weng and Moore were able to look beyond simple genetic proximity and ask which genes are functionally controlled by the DNA sequences carrying the trait-associated variants. Their analysis showed that these variants lie in regulatory regions that impact KLF1, a gene that acts as a central switch for red blood cell development. Experimental disruption of one of these regulatory regions reduced KLF1 activity, further supporting this connection. Together, these findings suggest that KLF1 is the most likely main gene through which genetic variation at this location affects red blood cell traits, although other genes may still play a secondary role. More broadly, this example illustrates how mapping regulatory DNA can clarify how genetic differences influence human traits by revealing which genes they control.

By combining dense regulatory maps with functional assays and human genetics, the UMass Chan-led ENCODE effort delivers a foundational resource for dissecting gene regulation, understanding developmental and cell specific programs, and clarifying how noncoding variation contributes to human disease.

The cCRE registry can be accessed online through an updated web portal.