From Genome to Diagnosis: Finding Answers for the Undiagnosed

Bringing answers to families through data-driven discovery and whole-genome analysis.

Rare diseases, typically defined as conditions that affect fewer than 1 in 2,000 people, are collectively not uncommon at all. Over 10,000 types of rare disease have been catalogued, and over 100 million people suffer from a rare disease worldwide. Most rare diseases are thought to have a genetic cause and typically exhibit symptoms and signs in childhood.

Patients with rare diseases and their families must often endure a long and frustrating journey to receive a genetic diagnosis, involving visits with multiple doctors, inconclusive test results, and clinical misdiagnoses. This diagnostic odyssey, as it is called, can last years. It represents a high emotional, physical, and financial burden to patients and their families, who desperately seek answers and a path to treatment. A patient’s diagnostic odyssey can only come to a close if the underlying genetic disorder has previously been discovered and characterized.

The lab of Ernest Turro specializes in finding the genetic underpinnings of previously unexplained rare diseases, allowing patients and their families to obtain a molecular diagnosis and bring their diagnostic odysseys to a close. Turro and his colleagues develop and apply statistical and computational methods to analyze large cohorts of whole-genome or whole-exome sequenced patients, along with detailed clinical information, to uncover the genetic causes of rare diseases.

“We study large heterogeneous collections of rare disease patients without a prior genetic diagnosis in the hope that certain subsets of those patients will turn out to share a common underlying disorder that can be identified statistically,” says Turro, PhD, Associate Professor in Genetics and Genomic Sciences and Arthur J. and Nellie Z. Cohen Professor of Pediatrics (Genetics) at the Icahn School of Medicine at Mount Sinai. “Understanding the genetic causes of rare diseases helps reveal the functions of genes and allows clinicians and families to reach clarity as what exactly caused a condition. Importantly, genetic diagnoses provide closure to those affected, helps delineate the clinical course of these conditions and enables the formation of disease-specific patient groups to share knowledge between families.”

Before the human genome was fully sequenced, scientists relied on pedigree analysis of large families to determine the inheritance pattern of single-gene diseases. They would work out which parts of the genome might be shared between affected individuals and narrow down the search to a single gene that might be implicated in that disease. But identifying families with rare diseases — and getting enough DNA samples for analysis — can prove challenging.

Over the last decade or so, the general approach to gene discovery has vastly improved. Now, every single base of the genome can be sequenced multiple times for high fidelity and at low cost. Scientists like Turro leverage large cohorts of unrelated individuals to tease out the genetic causes of rare diseases. For example, the United Kingdom’s National Genomics Research Library (NGRL) contains one of the richest datasets in the world for rare disease. The NGRL contains the whole-genome sequences of tens of thousands of individuals who have a rare disease with an unidentified cause, along with the sequences some of their affected and unaffected relatives.

Even with whole-genome sequencing, around a quarter of patients with rare diseases still do not obtain a genetic diagnosis. For example, around 60 percent of individuals with neurodevelopmental disorders remain undiagnosed despite comprehensive genetic testing. Undiagnosed diseases can include yet-to-be-described disorders that have not been previously documented or rare variations of more common diseases.

Earlier this year, Turro led a project using whole-genome sequencing data from the NGRL that resulted in the identification of a new neurodevelopmental disorder, caused by mutations in a single gene. Individuals with variants in a gene called RNU4-2 suffer from intellectual disability, short stature, seizures, and other abnormalities from an early age.

“This is actually one of the most common single-gene causes for this kind of disorder ever discovered,” he says. “It accounts for about half a percent of all cases of neurodevelopmental disorder, which in practice means that there are many tens of thousands of patients worldwide.”

Turro lab member Daniel Greene compared the burden of rare genetic variants in 41,132 non-coding genes between 5,529 unrelated cases with intellectual disability and 46,401 unrelated controls. Non-coding genes do not code for proteins and are usually not considered when searching for causes of rare diseases. The analysis pointed to variants in RNU4-2, a gene only 141 units long that codes for a small RNA molecule that plays a crucial role in gene splicing, a basic biological function of cells.

says Turro. "I was fortunate to meet with one of the first moms to receive a diagnosis for her child in person, and it was very moving. She’s met many other parents with affected children now, and they have a Facebook group where they can share experiences and photos of their children."

The next step for Turro’s lab and their collaborators is to study the precise mechanism that leads these individual variants to have such drastic effects. Exploring the molecular mechanisms more closely may provide biological insights that could one day lead to targeted interventions. The study also emphasizes the importance of looking beyond coding regions for genetic causes of rare disease. In fact, the Turro lab has already discovered that mutations in a related non-coding gene called RNU2-2P cause a similar disorder to RNU4-2 syndrome, and they have already collected data on 22 patients.

In collaboration with pediatrician Felix Richter, MD, PhD, Turro leads another project, called NICUnet, that centers around whole-genome sequencing of neonatal fatalities. Unexplained deaths affect roughly 1 in 1,000 live births, and the NICUnet initiative aims to uncover the genetic determinants behind such rare and unfortunate events. Across six states, 15 neonatal intensive care units are participating in the study, with almost 50 families enrolled so far.

“Up to this point, there have only been about 250 cases of neonatal fatalities published in the literature, all of them from single-hospital studies,“ Turro says. “We’re trying to do something at a much larger scale that should eventually allow us to identify genetic commonalities between different babies, and hopefully discover new causes of neonatal death through this work.”

Their first case was a baby who died from a severe kidney disease at only two days old, from a young Ukrainian couple who had moved to the United States shortly after the Russian invasion. Genetic analysis conducted by Turro lab member Kayleigh Rutherford revealed two genetic mutations — one from each parent, including one that could not have been found by standard clinical genetic testing — that together led to the development of polycystic kidney disease in their child.

“Later, they did in vitro fertilization and, because of our study’s findings, were able to screen out both of those genetic variants,” Turro says. “They were able to get pregnant with confidence, knowing that those variants wouldn’t cause the same issue again, and have recently had a healthy baby.”