A recent study suggests that long-read sequencing could enhance the speed and effectiveness of diagnoses for genetic disorders, cutting the time needed for diagnosis down from years to a mere days— all in one test and at a lower expense.
Approximately one in ten individuals globally suffers from a rare genetic disease, with nearly half of these cases remaining undiagnosed, despite advancements in genetic technologies and testing. Access to testing does not guarantee a timely diagnosis; the process can extend over five years or longer. This delay can be critical, especially for children who may need urgent treatment.
The problem stems largely from the use of short-read sequencing in current clinical practices, which is unable to capture information from specific genome regions, potentially overlooking vital diagnostic clues. Researchers from UC Santa Cruz are advancing a promising alternative known as long-read sequencing, offering a more thorough dataset for identifying genetic variations, reducing the need for multiple specialized tests, and simplifying the process of diagnosing rare diseases.
The findings published in The American Journal of Human Genetics indicate that long-read sequencing could significantly enhance diagnostic rates and decrease the time to achieve a diagnosis, all through a single test and at much lower costs. This study was spearheaded by key members of the UCSC Genomics Institute, including Professor Benedict Paten, Associate Professor Karen Miga, and former postdoctoral researcher Jean Monlong.
“Diagnosing rare diseases has been a longstanding challenge, and having a sequencing technology that simplifies this process could be a tremendous advancement — that’s what we explored in this study,” explained Shloka Negi, a Ph.D. student at UC Santa Cruz and the lead author of the paper.
“Currently, the success rate of genetic sequencing for diagnoses is quite unsatisfactory,” Paten remarked. “A significant issue may be the incomplete sequencing methods employed in clinical environments. In this investigation, we tested the idea that more thorough long-read sequencing generates valuable information for genetic diagnoses. We were thrilled to find a wealth of additional potentially significant genetic variations and epigenetic signals in our subjects. Although it’s still early, this information holds great potential, and understanding it fully will take time for the scientific community.”
Identifying rare diseases
This research concentrated on rare monogenic diseases, which arise from problems in a single gene.
To diagnose genetic conditions, scientists analyze genetic material for variants—differences in genes that may disrupt their normal function. The standard methodology for identifying these variants is short-read sequencing, which reads sequences of about 150-250 genetic base pairs—combinations of adenine (A), cytosine (C), guanine (G), and thymine (T).
However, short-read sequencing has limitations, as it can overlook critical details in certain genome areas, especially patterns of base pairs longer than 250. Moreover, it cannot perform “phasing,” which identifies the inheritance of genetic variants from each parent. This information is particularly beneficial for genetic evaluations, especially when parental genetic data is unavailable.
On the other hand, long-read sequencing can read extensive DNA segments at once, thus avoiding gaps that might lead to missing essential insights regarding gene variations. Additionally, long-read sequencing offers direct phasing data and insights into methylation — a chemical process influencing gene activity and associated with various diseases.
“Long-read sequencing is proving to be significantly more effective in certain instances, and we are systematically demonstrating this,” Negi stated.
Advancing methodologies
Researchers at the UCSC Genomics Institute have a noteworthy history of innovation in long-read sequencing. They are continuously enhancing methods for optimal sequencing and analysis across a variety of health research applications. Many techniques developed by these researchers, including the first complete “telomere-to-telomere” genome reference, are now contributing to better patient care.
Paten and Miga’s labs collaborated with clinicians to assess 42 patients with rare diseases—some receiving diagnoses through short-read sequencing or other specialized evaluations, while others remained undiagnosed. In certain scenarios, parental genetic data was available, while in others, it was not.
The long-read sequencing for patients was conducted by the Miga Lab, employing nanopore sequencing—a method for long-read sequencing established at UCSC—to produce highly accurate, comprehensive reads of the patients’ genomes at approximately $1,000 per sample.
The genomic data underwent analysis through computational methods developed in Paten’s lab, which identified small and large variants, phasing data, and methylation data via a single process known as the Napu pipeline. Depending on the computer’s processing capabilities, the analysis generally takes about a day or less and costs $100.
Solving diagnostic challenges
Upon sequencing and analyzing the patient data, researchers discovered that long-read sequencing generated a more thorough dataset compared to the results attainable through short-read sequencing.
Long-read sequencing provided conclusive diagnoses for 11 out of the 42 patients studied. This method captured all insights available from short-read data and offered additional information such as rare variant candidates, long-range phasing, and methylation—all achieved through a single, economical, and expedited protocol.
The diagnosed cases included four instances of congenital adrenal hypoplasia (a rare disorder characterized by underdeveloped adrenal glands that do not function correctly). The specific gene implicated in this condition resides in a challenging region of the genome that short-read sequencing fails to adequately characterize, and the prevailing clinical tests are insufficient.
“Long-read sequencing is likely the next optimal solution for unresolved cases featuring intriguing variants in a single gene or clear phenotypes,” Negi mentioned. “It can function as a unified diagnostic test, minimizing the need for multiple clinical visits and compressing what once took years into a matter of hours.”
On average, each patient had 280 genes, including several Mendelian disease genes connected to inherited disorders stemming from single-gene mutations, with significant protein-coding regions uniquely accessed through long reads that short reads missed.
“Long reads can unlock a significantly larger portion of the genome,” Negi stated. “However, it will take time before we can fully understand the new information that long reads expose. Many insights remained unseen in our clinical databases, which were constructed using short-read data mapping to a standard reference. Our study revealed that long reads can access around 5.8% more of the telomere-to-telomere genome that was simply unavailable through short reads.”
Other UC Santa Cruz researchers contributing to this study include Brandy McNulty, Ivo Violich, Joshua Gardner, Todd Hillaker, and Sara O’Rourke.
This research received partial funding from the Chan Zuckerberg Initiative.
