Almost every gene in a human body exists twice. One copy came from the mother, one from the father, and the two are not required to behave alike. One may be switched on while its twin stays quiet. When that uneven rule breaks down, serious illness can follow. For a long time it was hard to see these copy-specific differences at all, and harder still to tell which of them mattered. A team of geneticists now says it has an instrument for the job, University of Utah Health reported in an announcement reposted on the university’s news site.
The instrument is called FIRE, short for Fiber-seq inferred regulatory elements. It combines machine learning with advanced DNA sequencing to separate how the maternal and paternal chromosomes are regulated, and to find the regions that gene function depends on. The work appeared on Oct. 15 in Cell under the title “Somatic epimutations cap genetic determinism in the human diploid chromatin epigenome”. The authors come from the University of Utah, the University of Washington, Stanford University and Mayo Clinic, and an earlier version had been available as a preprint since June 2024.
“Only about 1 or 2% of our genome codes for proteins,” said Mitchell R. Vollger, assistant professor of human genetics at University of Utah Health and co-lead author. “This gives us a much sharper lens on the other 98%, which includes the switches that turn our genes on and off.”
The unmapped provinces #
Here is one way to picture it, offered as a metaphor and nothing more. The genome is like an old empire whose capital is small and thoroughly charted, the protein-coding genes, while most of its territory is administration: switches, relays, offices that decide when each order from the capital is carried out. Historians of empires know the provinces are where policy actually lives or dies. Geneticists have long had incomplete maps of theirs. The paper’s own summary says FIRE gives a fuller, more quantitative picture of accessible chromatin across the 6-billion-base-pair diploid genome than existing catalogs of regulatory elements, and that those catalogs carry biases nobody had noticed.
The tool turned up tens of thousands of DNA regions that earlier work had missed and that are likely to matter for gene regulation. These regions are more likely than the rest of the genome to hold DNA changes associated with human disease. That suggests they do important things for health that no one has yet described.
One letter among thousands #
Then there was celiac disease. Earlier studies had linked a particular genetic difference to the risk of the disease, but nobody knew whether it changed how genes work or simply sat near the real culprit, an innocent bystander. For anyone hoping to build genetic medicines, that difference matters.
The researchers found that the variant sits in a gene switch next to an immune gene that has also been linked to celiac disease. The change disrupts a site where gene-controlling proteins normally attach. In lab experiments, altering that single DNA letter was enough to change how open the switch was.
One letter. Thousands of nearby genetic differences, and this is likely the only one that changes gene regulation.
According to the paper’s highlights, the HLA region holds the densest parent-specific chromatin in immune cells, and in that region the team pinpointed specific binding events by gene-controlling proteins that disease-linked variants disrupt.
Where inheritance stops #
With FIRE the team could search on an unprecedented scale for stretches of DNA regulated differently on the mother’s and the father’s chromosome. Usually such a difference could be traced to a difference in the DNA itself. But in a large minority of cases the two copies were genetically identical at that spot and were still governed differently.
“About 40% of the time that we see a regulatory difference between the copies we inherited from mom and dad, there’s no genetic difference between them at that spot.”
Vollger called finding “the limits of how much our genetics can determine our gene regulation” the most surprising result of the work. These differences persist, too: they are passed on as cells divide. The paper calls them somatic epimutations, chromatin changes that stay stable through cell division, are spread by clonal expansion, and are not fixed by genetic sequence. What most of them do is unknown. The researchers read them as a new dimension of biological variability that genetics does not dictate. The paper also reports imprinted elements without the usual parent-of-origin differences in DNA methylation, and elements near gene starts that let genes escape silencing on the inactive X chromosome.
A reflection, not a finding: every system of inheritance, in politics as in biology, promises that what is handed down determines what follows. Here, in four cases out of ten, two identical inheritances are ruled differently, and nobody yet knows why.
The machine and the eye #
The method starts from an existing technique that marks the physically accessible parts of DNA. Advanced sequencing then reads long stretches of individual DNA molecules, and because each read is so long, small sequence differences usually show whether it came from the mother’s copy or the father’s. A marked region can point to an active gene or an important regulatory site, but it can also be an unimportant stretch that happened to be exposed to the marker at that moment. FIRE’s machine-learning layer checks each marked region and judges which are likely to matter. A researcher can do this by hand for a single site. The tool does it for billions of marked regions across the genome.
“Machine learning here is really copying something that we’re able to see with our own eyes in the data,” Vollger said. “But it’s doing it at a scale that we could never manage by eye.”
From variant to treatment #
Genetic tests can already explain, as one family learned from a saliva test at 33, why a disease runs through generations. Vollger’s hope is that linking changes in the genetic code to working differences in gene regulation will help show which variants actually cause disease. He sees that as an early step toward therapies that edit DNA to treat, or even cure, hereditary conditions.
“This might show us how we could edit the DNA switches that control a person’s genes to help lower their risk for a common disease, or even cure them of a rare disease,” he said. “It’s a small step toward turning the genetic variants we find into something doctors can act on.”
The research was funded by the National Institutes of Health, including the National Human Genome Research Institute, the National Institute on Drug Abuse, the National Institute of Allergy and Infectious Diseases and the National Institute of General Medical Sciences, and by the Brotman Baty Institute, Google.org and the Crohn’s and Colitis Foundation. The paper is open access and belongs to Cell Press’s Somatic Mosaicism across Human Tissues Network collection.