Gregor Mendel never intended to upend biology. In the 1860s, the Augustinian friar tended carefully cultivated rows of pea plants in his monastery garden in Brno — in what is now the Czech Republic — meticulously tracking the transmission of traits like seed color and pod shape across generations. When he published his findings in 1866, Mendel described an elegant, rule-governed system: traits are carried by discrete heritable units (later called genes), those units come in pairs, and they separate and recombine according to predictable mathematical ratios. His laws of segregation and independent assortment became the bedrock of modern genetics.
For more than a century, Mendel’s framework has guided scientists seeking to understand how traits — from eye color to cancer predisposition — pass from parent to child. But biology, as it turns out, has always kept secrets.
A sweeping new study published May 20 in Nature Genetics by researchers at Johns Hopkins University and Texas A&M University now reveals that a significant portion of heritable biological information travels by routes Mendel never imagined. Using cutting-edge long-read DNA sequencing technology, the team tracked how chemical modifications to DNA — a type of epigenetic mark known as DNA methylation — are passed across three generations of mice. Their conclusion: roughly 7% of the methylation patterns on non-sex chromosomes are inherited in ways that defy Mendel’s laws. And some of those deviations are genuinely, unexpectedly strange.
According to Andrew Feinberg, M.D., Johns Hopkins University, “Non-Mendelian patterns of inheriting epigenetics could be a faster way to acquire diverse or new traits than alterations in the genomic sequence itself, especially in response to environmental pressures.”
DNA methylation is one of the best-studied mechanisms of epigenetics — the field concerned with heritable biological changes that don’t involve alterations to the DNA sequence itself. When methyl groups (small clusters of carbon and hydrogen atoms) attach to specific sites along the genome, they can switch genes on or off, influencing everything from cell identity to disease susceptibility. Unlike the genetic code itself, these marks were long assumed to be largely reset between generations. The new study suggests that reset is far from complete.
A Genome-Wide Map of Inheritance
To conduct the study, lead author Adam Davidovich and colleagues at Johns Hopkins developed new experimental and computational methods for simultaneously reading genetic sequences and methylation patterns on the same DNA molecules — a technical feat made possible by long-read Oxford Nanopore Technologies (ONT) sequencing, which can decode stretches of DNA thousands to millions of letters long. They applied this approach to liver and muscle tissue from two genetically distinct strains of laboratory mice and their offspring across three generations: 26 animals in the first (inbred) generation, 34 in the second (F1 hybrids), and 19 in the third (F2 crosses).
The researchers searched for 12 distinct patterns of epigenetic inheritance across roughly 12 million autosomal DNA methylation sites, cataloguing both Mendelian and non-Mendelian patterns in both tissues. The majority — about 93% — followed Mendel’s rules, largely driven by so-called cis-acting methylation quantitative trait loci (meQTLs): genetic variants that control methylation of nearby DNA on the same chromosome. That part of the story was expected.
What was not expected was the breadth and variety of the exceptions.
Among the 522 non-Mendelian instances the team identified, 54 represented what the researchers call “emergent” epigenetic inheritance patterns — methylation states in offspring that were present in neither parent. Cross two mice with no methylation at a particular gene, and the offspring might somehow acquire it on both copies of that gene.
“The methylation seemingly appeared out of nowhere,” said Feinberg.
The team also discovered five genes not previously known to be imprinted — meaning their methylation pattern distinguishes which parent passed them down. Genomic imprinting, the most established form of non-Mendelian epigenetic inheritance, had already carved out a recognized exception to Mendel’s framework: certain genes are silenced not based on dominance or recessiveness, but based solely on whether they were inherited from the mother or father. Classic examples include the genes controlling growth factor IGF2 and the neurodevelopmental disorder Angelman syndrome. Finding five new candidate imprinted genes — including one on the X chromosome — adds meaningful new entries to that catalog.
But perhaps the most striking finding was the identification of a naturally occurring paramutation in a mammalian genome — the first of its kind ever reported outside of engineered or transgenic animals. Paramutation, a phenomenon first described in maize by corn geneticist R.A. Brink in the 1950s and subsequently observed in flies, occurs when the methylation state of one chromosomal allele is permanently altered by the presence of its partner allele — and that altered state is then passed to the next generation. It is, in a sense, a form of epigenetic contagion between chromosomes. The new study found it occurring spontaneously in the gene Capn11, which encodes a calcium-dependent protease critical to sperm development. In humans, reduced activity of the equivalent gene has been linked to infertility and azoospermia.
“It’s almost like the methylation is transferred to another allele,” concluded Feinberg.
The researchers traced the paramutation to a region of Capn11 overlapping a type of repetitive DNA element called an intracisternal A particle (IAP) — a subclass of endogenous retroviruses embedded in the genome. IAP elements are known to be sensitive to environmental exposures including diet and stress, and are among the few genomic regions that can escape the wave of epigenetic reprogramming that typically wipes methylation marks during embryonic development. Two additional highly likely paramutation events were also identified at related IAP sites near the genes Vps37c and a second Capn11 region.
The study also turned up 305 regions of sex-specific DNA methylation — nearly all confined to the liver and almost all showing higher methylation in females than in males. These patterns were absent in muscle, consistent with previous research documenting sex-biased gene expression and methylation in the human and mouse liver, likely tied to sex hormone signaling. Of the 233 genes associated with these liver-specific sex methylation patterns, 78% showed higher expression in males — the inverse of the methylation pattern, which fits with methylation’s general role in gene silencing.
Additionally, by examining X chromosome methylation in female mice, the team was able to map the extent of skewed X-chromosome inactivation — a process by which females silence one of their two X chromosomes in each cell. They confirmed a strong genetic bias in these mouse strains toward inactivating one parental X copy, and identified a novel imprinted site on the X chromosome in the gene Zfp92, which is implicated in X-linked intellectual disability. Analyses of Turner syndrome and autism spectrum disorder have long suggested an imprinted X-linked locus affecting cognition; this may be a first lead toward its identification.
What It Means for Genetics and Medicine
Taken together, the findings raise important questions about how much heritable biological information standard genetic analyses are missing. Genome-wide association studies (GWAS), which scan the genome for DNA variants linked to disease, typically ignore epigenetic marks entirely — yet those marks can regulate whether implicated genes are active or silenced. A family history of a disease may encode not just DNA mutations but heritable methylation patterns, some of which don’t follow the same rules geneticists have relied on for over a century. This could partly explain phenomena like incomplete penetrance — why some individuals who carry a disease-causing mutation never actually develop the disease — and why identical twins with the same DNA can have strikingly different health outcomes.
“This work may convince scientists to integrate both genomics and epigenomics more often for a complete understanding of how traits that produce disease and healthy states are inherited,” said Kasper Hansen, Ph.D., Johns Hopkins Bloomberg School of Public Health.
The emergent methylation patterns are particularly intriguing from an evolutionary standpoint. The study’s authors point to the possibility that these de novo epigenetic changes could fuel rapid phenotypic diversification in hybrid populations — potentially contributing to hybrid vigor, the well-documented but mechanistically murky phenomenon in which crossbred offspring outperform their parents. More speculatively, they suggest these patterns could provide a molecular mechanism for genetic assimilation, a concept first proposed by developmental biologist C.H. Waddington in the 1940s: the idea that environmentally induced traits can eventually become genetically fixed. Since methylated cytosines are prone to mutate at rates up to three times higher than unmethylated ones, epigenetic marks acquired in response to environmental stress could, over time, leave permanent mutational imprints in the genome.
The study is not without limitations. It examined only two mouse strains in two tissues under stable laboratory conditions; different strains, tissues, and environments would almost certainly reveal additional non-Mendelian patterns. The team was also unable to phase methylation across roughly one-third of the genome where the two strains share identical sequence — leaving a substantial dark matter of potential epigenetic inheritance undetected. And while the mouse findings are compelling, the extent to which they translate to humans remains an open question, one the researchers say they intend to pursue.
That human work, when it comes, may change clinical genetics in ways difficult to fully anticipate. For patients with unexplained familial disease, for populations exposed to environmental toxins, for the children of trauma survivors — a fuller accounting of how epigenetic inheritance really works may prove as transformative as Mendel’s original insight. It took the world thirty years after Mendel published to recognize what he had found. The field may not have that long to absorb what comes next.
Endnotes
1. Davidovich, A., Cuomo, D., Su, H., et al. “Non-Mendelian inheritance of DNA methylation patterns in mice.” Nature Genetics (2026). https://doi.org/10.1038/s41588-026-02604-z
2. Johns Hopkins Medicine. “Lab study reveals patterns of inheritance that defy Mendel’s laws.” EurekAlert! (May 20, 2026). https://www.eurekalert.org/news-releases/1128940
3. Jablonka, E. & Raz, G. “Transgenerational epigenetic inheritance: prevalence, mechanisms, and implications for the study of heredity and evolution.” Quarterly Review of Biology 84, 131–176 (2009).
4. Reik, W. & Walter, J. “Genomic imprinting: parental influence on the genome.” Nature Reviews Genetics 2, 21–32 (2001).
5. Ferguson-Smith, A.C. “Genomic imprinting: the emergence of an epigenetic paradigm.” Nature Reviews Genetics 12, 565–575 (2011).
6. Chandler, V.L. “Paramutation: from maize to mice.” Cell 128, 641–645 (2007).
7. Hollick, J.B. “Paramutation and related phenomena in diverse species.” Nature Reviews Genetics 18, 5–23 (2017).
8. Morgan, H.D., Sutherland, H.G., Martin, D.I. & Whitelaw, E. “Epigenetic inheritance at the agouti locus in the mouse.” Nature Genetics 23, 314–318 (1999).
9. Cockett, N.E. et al. “Polar overdominance at the ovine callipyge locus.” Science 273, 236–238 (1996).
10. Malcher, A. et al. “Potential biomarkers of nonobstructive azoospermia identified in microarray gene expression analysis.” Fertility and Sterility 100, 1686–1694 (2013).
11. Lane, N. et al. “Resistance of IAPs to methylation reprogramming may provide a mechanism for epigenetic inheritance in the mouse.” Genesis 35, 88–93 (2003).
12. Cooper, D.N. et al. “Where genotype is not predictive of phenotype: towards an understanding of the molecular basis of reduced penetrance in human inherited disease.” Human Genetics 132, 1077–1130 (2013).
13. Feinberg, A.P. “The key role of epigenetics in human disease prevention and mitigation.” New England Journal of Medicine 378, 1323–1334 (2018).
14. Skuse, D.H. et al. “Evidence from Turner’s syndrome of an imprinted X-linked locus affecting cognitive function.” Nature 387, 705–708 (1997).
COPY II (2-3 PARAGRAPHS)
IMAGE CREDIT: NASA.
Leave a Reply