Connecting cases to uncover an ultrarare neurological disorder

The Pareto Principle states that 80% of consequences come from just 20% of causes. This 80/20 “rule” is often discussed regarding business or quality control, but the concept can also be seen in diseases where a large portion of cases are caused by just a handful of identified reasons. 

In 2024, researchers at St. Jude scoured the genomes of patients with unresolved pediatric epilepsies. While half of pediatric epilepsies are currently diagnosable, most are linked to just 27 genes. The patients with no diagnosis were the outliers of the Pareto Principle — the other half.

Caused by a wide range of seemingly arbitrary genetic alterations, these patients exist in a state of unknown — individually considered “ultrarare,” but collectively represent a dire need for answers. And answers are what the team, led by Heather Mefford, MD, PhD, St. Jude Center for Pediatric Neurological Disease Research, Department of Cell & Molecular Biology, and Department of Genomic & Translational Neuroscience, hoped to find. 

The researchers used a screening technique that looked at chemical changes to DNA called methylation marks. Methylation is a process that allows cells to control how genes are accessed. Across a genome, the process usually follows a predictable rhythm. Yet, while looking at methylation patterns within their genomic dataset, the researchers noticed a region in one patient’s genome that had fallen out of sync.

“Once we found an abnormally methylated region in this patient’s genome, we identified an underlying change that was shutting down expression of BCLAF3, a gene on the X chromosome that we don’t know much about,” Mefford said. “All the features in this patient and the characteristics of the genomic change eventually led us to believe this abnormality was likely causing the epilepsy.”

Repeat expansion disrupts gene expression

The abnormality the researchers spotted is called a tandem repeat expansion, where a short strand of DNA is replicated hundreds, or even thousands, of times in a row. In this patient, the repeat expansion appeared right beside BCLAF3 and was being methylated repeatedly, drastically altering how the gene was being expressed.

This phenomenon is very similar to Fragile X syndrome, one of the most common causes of intellectual disability in boys, in which a repeat expansion on the X chromosome gets expanded and hypermethylated, shuts down gene expression, leading to disease. 

A single patient does not define a disorder, however. Mefford’s team knew they had to go further.

For the 50% of patients with undiagnosed ultrarare epilepsies, identifying even two individuals with overlapping genetic alterations requires searching across multiple datasets containing thousands of patient samples. Mefford reached out to collaborators with methylation arrays and research groups with large whole-genome sequencing datasets with a simple question: “Are you willing to look at your dataset and see if anybody has a similar pattern on the X chromosome?”

Searching the world for an ultrarare disease

Many groups answered this call, culminating in 12,000 methylation profiles and nearly 16,000 short-read patient genomes searched globally. From this exhaustive dataset, five affected males were identified where the BCLAF3 expansion was identified as a likely contributor to their condition. However, even in this highly selected cohort, surprises appeared. Most notably, two of the boys were cousins, and each had a separate, different genetic condition — one boy had Williams syndrome, caused by a deletion on chromosome seven, and the other had Fragile X syndrome.

“The family had undergone additional testing because the boys were more severely affected than they should have been for those conditions,” Mefford explained. “We found the BCLAF3 repeat expansion in both of them. The increased severity of their conditions led us to believe that the BCLAF3 expansion is pathogenic.”

X chromosome phenomenon largely spares women

Why were these cases all male? Women have two X chromosomes, while men only have one. In women, each cell randomly inactivates one of their X chromosomes. It might be expected that female carriers of an X-linked disorder, such as the children’s mothers, would exhibit at least 50% of the disease effect. However, decades of research have shown that this is often not the case. 

One reason is a phenomenon called skewed X-inactivation, in which one X chromosome is preferentially inactivated far more often than the other, leading to an uneven expression of disease-causing variants. Indeed, sequencing from three of the boys’ mothers revealed they all had skewed X-inactivation.

“When we examined the mothers’ methylation profiles, the X chromosome that was consistently inactivated was always the one carrying the BCLAF3 repeat expansion. This means they might have shown some disease effects if X inactivation had occurred randomly,” Mefford said. “Skewed X inactivation is common in females carrying X-linked conditions and can be protective because it preferentially silences the chromosome carrying the disease-causing mutation.”

No disease too rare, no family left alone

The findings of the study were published in 2026 in Genome Medicine, but the researchers are continuing to connect the dots between BCLAF3 repeat expansion and neurodevelopmental disorders. The real impact of understanding ultrarare variants, however, lies beyond one gene. Each newly identified variant is one more disease contributor for which clinics can screen; each unique natural history reveals more about how these diseases manifest and even work in tandem; and each family who can identify their variant can better understand it, manage it and feel less alone.

“Many of the technologies that we employed in this investigation are already being used in the clinic to diagnose patients with neurodevelopmental and neurologic conditions much earlier,” Mefford said. “We still predict these conditions are going to be rare given the number of patients we’ve looked at through these collaborations. But it’s still important for every single patient who is identified.”