A Prevalent Recessive Neurodevelopmental Disorder Hidden in the Noncoding Genome
Tachyon's Co-Founder and Managing Partner, Ernest Turro, PhD, Associate Professor of Genetics and Genomic Sciences at Mount Sinai's Icahn School of Medicine in New York, has just published a major new paper in Nature Genetics that identifies a previously unknown recessive neurodevelopmental disorder — the most prevalent of its kind ever discovered.
"Our discovery will enable tens of thousands of families affected by this previously hidden genetic condition to receive closure through a genetic diagnosis. Understanding that the disorder stems from a loss of U2-2 RNA points to potential gene replacement strategies in the future." — Ernest Turro, PhD
This finding is the latest in a series of landmark discoveries from Ernest's research group. In 2023, the team published a Nature Medicine paper identifying 19 new candidate genetic causes of rare diseases, of which, at that time, they investigated and validated three in detail. In a separate Nature Medicine paper published in 2024, they identified mutations in the gene RNU4-2 as the cause of the most prevalent autosomal dominant neurodevelopmental disorder known to date, now known as ReNU syndrome. The following year, in a 2025 Nature Genetics paper, they showed that mutations in a related gene, RNU2-2, cause a severe dominant condition with prominent epilepsy. This latest Nature Genetics paper adds another chapter to the story — demonstrating that different variants in RNU2-2 cause a distinct recessive neurodevelopmental disorder that is, by a significant margin, the most prevalent such disorder known to date.
Why rare diseases matter — and why most remain unsolved
Rare diseases are individually uncommon but collectively affect roughly 1 in 20 people. Over 10,000 such conditions have been described, yet fewer than half have been tied to a specific genetic cause. For the families affected, this means years — sometimes decades — without a diagnosis, without an explanation for their child's symptoms, and without a roadmap for care. Identifying causal relationships between genetic variation and rare diseases enables the delineation of distinct subgroups of individuals sharing the same pathology, provides the foundational knowledge required to issue genetic diagnoses to patients going forwards, reveals important molecular biology about the genome, and opens up opportunities for developing targeted therapies.
The human genome contains roughly 20,000 protein-coding genes and tens of thousands of noncoding genes — genes that produce functional RNA molecules that are not translated into proteins. While protein-coding genes have been the focus of most rare disease genetics research, noncoding genes have been less studied, in part because the tools to analyze them at scale have only recently matured. Some of the remaining unsolved rare diseases may be caused by variants in these overlooked regions of the genome.
Ernest's research program is built on the premise that large-scale datasets of sequenced genomes from rare disease patients, combined with sophisticated statistical methods, can uncover many of the causes that remain unknown, both coding and noncoding.
What the new study found
Using data from over 110,000 participants in the UK's National Genomic Research Library, the team applied their BeviMed statistical method and found very strong evidence for a recessive neurodevelopmental disorder caused by biallelic (two-copy) variants in RNU2-2. They identified more than 30 affected families in the UK dataset and replicated the finding across independent collections in the United States, Italy, and the Netherlands. RNU2-2 is a noncoding gene with a length of only 191 base pairs that is transcribed into U2-2 small nuclear RNA. The gene’s function is not known, but its sequence is almost identical to that of a different gene, RNU2-1, that encodes a well-known component of the spliceosome, the cellular machinery that processes gene instructions before they can be translated into proteins.Children with this condition — now called recessive ReNU2 syndrome — typically present in infancy or early childhood with developmental delays, low muscle tone, and limited or absent speech. Many develop seizures. The clinical picture ranges from mild learning difficulties to severe progressive epileptic encephalopathy, depending on which specific variants a child carries.
RNA sequencing of blood samples from patients revealed the biological consequence of the variants: affected individuals showed a mean reduction in U2-2 RNA expression of over 98% compared to controls, probably because the disease-causing variants destabilize the U2-2 RNA. Observations in unaffected “carriers” of these variants, such as healthy parents, demonstrated that different disease-causing variants in RNU2-2 destabilize the U2-2 RNA to different degrees, and the severity of disease appears to be inversely related to the reduction in U2-2 abundance. In the parlance of human genetics, the variants alleles are said to be “hypomorphic.” Parents, who have only one “bad” copy of the gene, appear to be unaffected in part because their cells respond by increasing expression of the remaining “healthy” copy of the gene. It may be that individuals with borderline low U2-2 abundance caused by the presence of two modestly hypomorphic alleles have intermediate phenotypes (e.g., mild autism or intellectual disability but no seizures). If so, there may exist a mild form of the condition that is relatively common but not represented in the rare disease dataset used to make the discovery, as it tends to include severely rather than mildly affected individuals. Importantly, the revelation of a loss-of-expression mechanism opens the door to diagnostic RNA-based blood tests and, potentially, gene replacement therapies.
Why this matters
Recessive ReNU2 syndrome is the most prevalent recessive neurodevelopmental disorder ever identified, accounting for roughly 10% of all families with a known recessive neurodevelopmental disorder. It is around 60% as common as the dominant RNU4-2 disorder (ReNU syndrome) that Ernest’s group described in 2024 — unusual, because the most prevalent neurodevelopmental disorders are typically dominant. The high prevalence may reflect an elevated mutation rate in RNU genes.
Because this is a recessive condition — each parent may carry one altered copy without being affected — the discovery has direct implications for family planning through preconception counseling and prenatal genetic testing.
The computational approach that enabled this discovery — Bayesian association methods applied to genomic data from national sequencing programs — exemplifies a broader principle that we at Tachyon find compelling: that sophisticated computational methods applied to large, well-curated biological datasets can uncover previously hidden patterns. Ernest's academic work and Tachyon's investment thesis share a common conviction that the most important biological insights of the coming decade will be extracted computationally from data that already exists or can be generated at scale.
What comes next
The team has launched the INDEED study at Mount Sinai to enroll affected families — to deliver diagnoses, characterize the natural history of the condition, and understand how U2-2 RNA loss disrupts neurodevelopment. The clear loss-of-expression mechanism points naturally toward gene replacement strategies. While a specific treatment does not yet exist, the identification of a defined molecular target is the critical first step.
A family foundation, the ReNU2 Syndrome Foundation, has been established to connect affected families and support research. For many of these families, this paper represents something that no amount of clinical care could previously provide: an answer.
The paper, "Biallelic variants in RNU2-2 cause the most prevalent known recessive neurodevelopmental disorder," was first-authored by Daniel Greene, PhD, co-authored by an international team of collaborators, and senior-authored by Ernest Turro, PhD. It is published in Nature Genetics.
