ME/CFS, long Covid, post-traumatic stress disorder, rheumatoid arthritis and multiple sclerosis share almost no individual risk genes, yet a University of East Anglia team reports that the genes involved in each converge on the same handful of biological networks. Prof Dmitry Pshezhetskiy of Norwich Medical School, who led the work, put it this way: “Although these conditions are triggered by completely different events, they may ultimately disrupt the same fundamental biological systems.” The analysis appeared in the Journal of Translational Medicine in early September 2026, and it drew immediate criticism from independent geneticists about its statistics and its proprietary tools.
The five conditions and the five shared systems
The set is exact and was counted from the paper’s own title. The paper, titled “Beyond genes: EpiSwitch and Orion platform-powered 3D genome architecture biomarkers reveal shared biology across ME/CFS, long COVID, PTSD, rheumatoid arthritis, and multiple sclerosis,” names these five and no others. Two are post-infectious or fatigue-defined illnesses, one is a trauma-related psychiatric condition, and two are autoimmune diseases, which is why the grouping looks odd on a clinical chart.
According to the University of East Anglia’s release, the analysis found convergence in five systems: immune and inflammatory signaling, mitochondrial energy production, metabolic regulation, stress-response mechanisms and neuroendocrine signaling. Pshezhetskiy described the result as “something approaching a biological unifying theory of fatigue.”
Genome folding and the EpiSwitch Orion platform
No new patients were enrolled. The team worked from published genome-wide association studies together with three-dimensional genome data from earlier ME/CFS studies, and applied the EpiSwitch Orion platform from Oxford BioDynamics to examine how DNA folds so that regions far apart in the linear sequence can touch. Dr Ewan Hunter explained, in comments carried by News-Medical, that DNA is folded in cells, so distant stretches can sit side by side and regulate one another.
Individual genes showed minimal overlap among the conditions. The overlap appeared one level up, in how the genes interact in networks. Pshezhetskiy said that when the interactions were analyzed, “suddenly, the diseases appeared deeply connected.” A separate summary from ME Research UK adds that the genetic datasets included DecodeME, the largest DNA study of ME/CFS, and that the work identified central hub genes that might explain how different triggers produce similar exhaustion.
One of those hubs, LAG3, is linked to T-cell exhaustion, in which immune cells tire after prolonged activation. That connection to tired immune cells is the most concrete biological handle the paper offers, since T-cell exhaustion is a documented state in chronic infection and cancer, and it gives later laboratory work a specific molecule to test in ME/CFS patients rather than a diffuse network. The authors flagged it as a candidate in ME/CFS specifically. The partners on the project were Oxford BioDynamics, the London School of Hygiene and Tropical Medicine and Cornwall Partnership NHS Foundation Trust.
Edinburgh and King’s College London on the statistics
The Science Media Centre gathered reactions the day after the release. Dr Sjoerd Beentjes of the University of Edinburgh said the claim of a unifying theory appears overstated, noting the analysis lacks precise metrics for comparing how similar the conditions are and that the proprietary EpiSwitch tools prevent independent scrutiny.
Prof Chris Ponting, also of Edinburgh, pointed to permissive statistical filters. The study used a threshold of p below 0.01, where genome-wide work normally demands about 5 x 10^-8. He added that chromosome data from blood cells may not match the cell types that matter in disease, such as neurons for PTSD and ME/CFS, and that the methods are “locked within black boxes,” preventing reproduction.
Prof Carmine Pariante of King’s College London took a milder view, calling the work “largely confirmatory” because it reuses existing datasets and does not identify new mechanisms, though he said the compiled evidence may help researchers and patients understand immune involvement across the conditions. His assessment sits between the authors’ language of a unifying theory and the Edinburgh critique of overreach, and none of the three commentators disputed that the five conditions were the ones analyzed.
Hub genes, blood tests and the unproven leap
The authors themselves describe the hub genes as candidates needing confirmation before clinical conclusions are drawn, and the ME Research UK summary labels the work hypothesis-generating. Coverage in Labcompare reports that the team hopes the shared signatures could support objective blood tests and treatments effective across several chronic conditions at once. The UEA release raises the prospect of blood-based diagnostic tests and treatments that work across conditions, but no test or therapy was evaluated in the paper, and the release’s own hedge is that the shared signatures might one day support them.
The unresolved issue is the one Ponting raised: whether the same network overlap survives when the analysis is repeated with stringent thresholds, in disease-relevant cell types, and with methods outside investigators can run.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
- Common allergy, bladder and sleep pills tied to sharply higher dementia odds
- The second-largest U.S. reservoir just fell to its lowest level ever recorded
- Security experts still urge phone owners to switch off one location-tracking setting
- The FBI warns a new phishing trick can hijack your account without your password
