Long-Read Sequencing Detects Rare Disease Variants

Rare diseases are defined as disorders that affect fewer than 1 in 2,000 people. They often remain undiagnosed or even misdiagnosed, owing to the vast diversity of disorders and relatively common symptoms that can mask the underlying disease.

Approximately 80% of rare diseases have a genetic cause. As a result, diagnosis often involves genetic testing consisting of clinical short-read exome and genome sequencing approaches. While these approaches have positively impacted rare disease diagnostics, there remain technical limitations when using these techniques to detect disease-associated variation in complex regions of the genome.

A study published in the American Journal of Human Genetics has demonstrated the impact of HiFi long-read sequencing (LRS) in identifying rare disease variants. These findings highlight the potential of HiFi LRS technology to replace multiple diagnostic tests with a single test capable of accurately identifying all types of clinically relevant disease variants.

Limitations in existing rare disease diagnostic tests

Contrary to definition, the disease burden of rare diseases is collectively high with an estimated 25 to 30 million Americans affected across 7,000 conditions. Alarmingly, the average time for an accurate diagnosis is 4–8 years; with around 30% of children with a rare disease dying before they reach 5 years old. These statistics are often attributed to the limitations of current genetic testing methods.

“Based on rare disease research studies, including our paper, we now see that short-read technologies show limitations for certain variant types, predominantly structural variants (SVs) and repeat expansions,” Alexander Hoischen, professor of genomic technologies for immune-mediated and infectious diseases at Radboud University Medical Center, told Technology Networks.

“Like many other human genetic laboratories, Radboud currently still uses multiple orthogonal test modalities to complement short-read exomes and genomes. Our recent work has shown that short-read genome sequencing cannot fully replace all other tests – hence truly generic testing is not yet feasible. Our paper shows how we see great potential for HiFi genomes as a truly generic first tier test for rare diseases.”

To test the ability of LRS technology to overcome these challenges, the researchers analyzed 100 patient samples with 145 clinically-relevant variants. The samples were chosen as the genetic cause of the rare diseases had been difficult to identify in previous investigations using short reads combined with various supplementary tests.

HiFi LRS was able to identify 93% of challenging pathogenic variants from the researcher's dataset. The technology used was also able to detect genetic variations missed by short-read approaches, including ∼90% of SVs, single nucleotide variants/insertions or deletions (indels) in homologous sequences and expansions of short tandem repeats.

“This retrospective study was conducted on some of the most difficult variants a diagnostic laboratory can face,” Christian Gilissen, professor of genome bioinformatics at Radboud University Medical Center, told Technology Networks. “Our results suggest that HiFi genomes may be the best first-tier test, potentially replacing many – if not all – standard-of-care test modalities. Our study, however, also highlights a few areas where sequencing, variant calling and variant visualizations can still be further improved.”

“The following variant types require additional efforts: repeat expansions in very long 'AG-rich' repeats and complex structural variants including balanced SVs with breakpoints in or near the centromeres,” Gilissen explained.

The future of rare disease diagnostics

This study demonstrates the diagnostic power of long reads, and the researchers are continuing their efforts to bring this approach into clinical practice.

“We really have the vision of HiFi long-read genomes as a first-tier generic germline test for all rare diseases,” said Gilissen. “This still requires an additional step in scalability, throughput and cost. However, we are just finishing a prospectively designed clinical utility study consisting of 1,000 HiFi genomes that strongly supports the notion of long-read sequencing as a comprehensive first tier test.”

Beyond rare disease diagnoses, continued advancements in technology could also accelerate the use of LRS in other research areas. “Population genomics and more complete reference genomes also benefit from long-read genomes. As such, also more common diseases may soon be able to utilize long-read benefits,” said Hoischen. “In addition, pharmacogenomics – the study of how complex genes impact drug response – is also an area that will potentially benefit. But our research, for now, will remain strongly focused on rare disease applications.”

Reference: Höps W, Weiss MM, Derks R, et al. HiFi long-read genomes for difficult-to-detect, clinically relevant variants. AJHG. 2025;112(2):450-456. doi: 10.1016/j.ajhg.2024.12.013

About the interviewees

Professor Alexander Hoischen heads the research group “Genomic Technologies for Immune-Mediated and Infectious Diseases” at Radboud University Medical Center, which builds expertise in the identification of rare disease genes using the latest genomics tools.

Hoischen co-leads a work package in the EU-funded H2020 project “solving the unsolved rare diseases (SOLVE-RD)”, which will be followed up by his work package-lead in the newly funded European Rare Diseases Research Alliance (ERDERA) project, both with a strong focus on long-read sequencing efforts in the rare disease space. Most recently Hoischen has received a Vici grant for the project “solving enigmas of undiagnosed inborn errors of immunity (SOLVE-IEI)”.

Hoischen and his local, national and international network of colleagues and collaborators will continue to pioneer novel and disruptive technologies that allow new scientific insights and rapid translation into clinical and diagnostic practice.

Christian Gilissen is a professor of genome bioinformatics at Radboud University Medical Center. Gilissen’s group works on the development and application of computational methods for the analysis of high-throughput sequencing data, the interpretation of genome variation, the identification of genetic causes of rare diseases and studies on the underlying biology of (de novo) mutations. His work has been published in various high-ranking genetics journals such as Nature, Nature Genetics and Nature Neuroscience.