Scientists on the Salk Institute and the Arc Institute, together with their collaborators, unveiled the primary body-wide single-cell atlas of two main epigenetic programs: three-dimensional genome folding and DNA methylation, measured concurrently in the identical cells.
The atlas spans 86,689 cells from 16 human tissues, revealing 35 main cell varieties and 206 subtypes, and is freely out there on-line. The work is a part of the Nationwide Institutes of Well being’s 4D Nucleome (NIH 4DN) program, which goals to know how the genome is organized in house and time to manage gene expression in well being and illness.
As a result of the 2 epigenetic layers had been measured collectively, the researchers might evaluate what every layer says a few cell’s identification. And whereas usually the 2 footage agree, they discovered that generally they don’t.
The Salk paper “Human physique single-cell atlas of 3D genome group and DNA methylation” was printed alongside 5 different NIH 4DN papers in Science, and three others in Science Advances.
The Human Genome Venture, accomplished in 2003, produced a linear learn of the three billion DNA letters within the human physique. However the letters alone don’t clarify how a single genome produces lots of of various cell varieties. That info lives within the epigenome within the type of chemical modifications and structural folds layered on high of the DNA sequence, the place they’ll change genes “on” and “off” in patterns particular to every cell sort.
![Caption: Salk scientists Jingtian Zhou (left), Jesse Dixon (center), and Joseph Ecker (right) profiled 86,689 cells across 16 human tissues, linking cell-type-specific epigenetic features to disease risk and revealing that a cell’s 3D genome and DNA methylation don’t always tell the same story. [Salk Institute]](https://www.genengnews.com/wp-content/uploads/2026/07/260723-pr-ecker-dixon-authors-232x300.jpg)
Two of probably the most consequential epigenetic options are 1) DNA methylation, the place small chemical teams referred to as methyl teams are hooked up to particular DNA bases, and a couple of) 3D genome group, the place intricate loops, folds, and compartments carry distant stretches of DNA into contact. Each affect gene expression, however that they had by no means been measured collectively in single cells throughout the human physique.
“There was an appreciation for attempting to know, on the particular person cell degree, how the genome is organized, in order that we are able to get a greater concept of how genetic variants impression illness,” mentioned co-corresponding creator Joseph Ecker, PhD, a professor and Salk Worldwide Council Chair in Genetics at Salk and a Howard Hughes Medical Institute investigator. “Some cell varieties could also be extra susceptible than others to genetic variants, as a result of the genome is organized otherwise in numerous cell varieties—and whether or not a variant issues can rely upon that group.”
Why is noncoding DNA related in illness?
Most disease-associated genetic variants fall within the noncoding areas of the genome. That has made it troublesome to determine how a variant contributes to illness, which cell sort it acts in, and what gene it in the end impacts.
The brand new atlas identifies greater than 1.36 million differentially methylated areas and 283,606 differential chromatin loops throughout the human physique’s cell varieties, utilizing tissues from the center, mind, lungs, abdomen, pores and skin, and extra. When the researchers overlaid genetic variants identified to boost illness threat, particular pairings emerged like variants for blood-glucose regulation concentrated in endocrine cells, atrial fibrillation variants in coronary heart muscle cells, balding variants in pores and skin fibroblasts, and bipolar dysfunction and schizophrenia variants in excitatory and inhibitory neurons.
“A whole lot of the genetic variation that predisposes somebody to illness is in noncoding components of the genome,” mentioned co-corresponding creator Jesse Dixon, MD, PhD, affiliate professor and Helen McLoraine Developmental Chair at Salk. “By including within the 3D genome facet, we are able to doubtlessly bridge that hole—connecting noncoding variations with the genes they have an effect on in particular cells and tissues.”
What occurs when two epigenetic lenses disagree?
One of many research’s most shocking findings is that DNA methylation and 3D genome construction don’t all the time inform the identical story a few cell. In skeletal muscle, the group discovered fibers that seem like mature, differentiated muscle cells by their 3D genome folding, however nonetheless carry the methylation signature of muscle stem cells. The reverse virtually by no means occurs. Essentially the most believable rationalization, they defined, is that these cells are caught mid-differentiation, with 3D structure updating first and methylation catching up.
Comparable mismatches appeared in Schwann cells of the peripheral nervous system and in placental trophoblasts. The sample means that completely different epigenetic options replace on completely different time scales throughout cell state transitions—a discovering that would reshape how researchers outline “cell sort” in grownup tissues and the way they observe cells shifting between states in illness.
The atlas additionally revises a long-standing assumption about “non-CG methylation,” an uncommon type of methylation beforehand regarded as largely confined to mind cells and stem cells. The research exhibits that it carries cell-identity info throughout many human tissues, together with muscle, pancreas, and immune cell varieties, at decrease however biologically significant ranges.
“The inconsistency between modalities could also be additional used to find out what cell populations are switching between one another in grownup tissues and ailments, which might, for instance, increase our understanding of most cancers cell dynamics,” mentioned co-first and co-corresponding creator Jingtian Zhou, PhD, a former graduate researcher in Ecker’s lab who now leads his personal lab on the Arc Institute.
A public useful resource for scientists and synthetic intelligence
To make the atlas broadly usable, the group constructed an interactive internet browser that lets researchers visualize DNA methylation and 3D chromatin contacts throughout each tissue, cell sort, and subtype within the research. The underlying knowledge, together with 195 billion methylation measurements and 18 billion chromatin contacts, are freely out there.
The useful resource arrives as synthetic intelligence instruments are more and more used to foretell the practical impression of genetic variants. Atlases like this one can present the labeled, cell-type-resolved coaching knowledge that fashions must make correct predictions—a bottleneck that has traditionally restricted the sphere.
For instance, in a companion paper in the identical concern of Science, a research led by Bing Ren, PhD, from the New York Genome Middle and Columbia College used the atlas’ cross-tissue methylation knowledge to indicate {that a} substantial fraction of the mind’s resident immune cells, referred to as microglia, are changed by cells resembling blood monocytes between roughly ages 50 and 75. The discovering challenges the long-held view that microglia persist from embryonic improvement all through the life span.
“DNA methylation patterns are particular to every cell sort and analogous to a mobile barcode,” mentioned Ren, who additionally co-authored the Salk-led research. “The excellent cross-tissue DNA methylation atlases present that the ageing microglia within the human hippocampus extra carefully match the monocytes from peripheral blood than microglia from younger adults, offering a vital clue for the organic identification of those cells.”
The NIH 4D Nucleome consortium, of which this research is part, goals to increase this type of mapping into the fourth dimension: time. A 4D understanding of the genome—how its construction and chemistry change as cells develop, age, and reply to illness—stays a serious aim, and the cross-tissue atlas supplies reference scaffolding that future time-course research will construct on.
Together with scientists from the Salk Institute and Arc Institute, investigators from the College of California, San Diego, Columbia College, New York Genome Middle, College of California, Los Angeles, Harvard, Henan College in China, Vanderbilt College, Stanford College, Broad Institute, College of Sheffield within the U.Ok., Yale, College of Florida, College of Freiburg in Germany, College of Graz in Austria, and Nanchang College in China; and Chongyuan Luo additionally contributed to the Science paper.
