The diagnostic odyssey of rare diseases continues, affecting an estimated 300 million people worldwide, yet up to half of cases remain unsolved using traditional diagnostic approaches. In the wake of their partnership to tackle this growing bottleneck, we connected with Ryan Taft, Chief Scientific Officer at iHope Network/Genetic Alliance, and Stacy Musone, Head of Global Market Development at PacBio, to discuss the realities of rare disease diagnostics and the impact of advanced genomics.
From a diagnostic perspective, what remains the biggest challenge in rare disease care today?
Stacy Musone: The biggest challenge is that many patients remain undiagnosed despite extensive testing. Diagnostic testing often begins with targeted gene panels or exome sequencing, which look for specific variants linked to known conditions. If those tests are inconclusive, clinicians may move to short-read whole genome sequencing (WGS) to examine the entire genome.
In many cases, the problem is not that the genetic variant is absent, but that current technology cannot detect it clearly. Short-read WGS works by fragmenting and amplifying DNA, which can make it difficult to identify complex structural variants, repeat expansions, and epigenetic signatures that are often associated with rare diseases.
As a result, patients may undergo multiple tests, each capturing only part of the genetic picture and delaying a definitive diagnosis. Improving the completeness of genomic analysis and streamlining testing pathways could help shorten the diagnostic odyssey for patients with rare diseases.
How can pathologists better recognize when a rare disease should be suspected and trigger appropriate testing?
Ryan Taft: Rare genetic disease should be considered when a patient’s clinical presentation does not clearly fit a single organ system, when symptoms are unusually severe compared with common conditions, or when disease begins very early in life. Clues can include multisystem involvement, developmental regression, unexplained neurologic findings, repeated metabolic abnormalities, or unusually severe or atypical symptoms. A negative or inconclusive targeted test – especially after gene panel testing – should not rule out a genetic cause but should prompt further investigation.
Pathologists play an important role in identifying these cases and coordinating diagnostic information. By correlating histopathology, laboratory findings, and family history, they can help identify patterns that may not be immediately apparent. Careful documentation of patient phenotypes, ideally using standardized terminology, can also improve the chances that genomic testing will identify a diagnosis. Research suggests that the most helpful information is not the number of clinical features recorded, but the presence of precise, highly informative descriptions of the patient’s symptoms.
When a rare genetic disease is suspected, WGS is increasingly recommended as the most comprehensive single test. WGS can detect changes across both coding and noncoding regions of the genome, including structural variants, copy number changes, repeat expansions, and complex rearrangements. If WGS is not available, exome sequencing or targeted tests may still be used, but starting with less comprehensive testing can lead to multiple sequential tests, longer diagnostic delays, and higher overall costs.
What role do multidisciplinary teams play in shortening time to diagnosis?
RT: Rare diseases rarely fit within a single medical specialty, so faster diagnosis often depends on structured collaboration.
Medical geneticists and genetic counselors help refine and update the patient’s clinical phenotype. Pathologists contribute tissue-level findings, laboratory data, and biomarker insights that help interpret molecular results. Bioinformaticians manage genomic data analysis, including variant detection, quality control, inheritance modeling, and filtering of large datasets. When these roles operate in isolation, patients often experience prolonged delays in diagnosis.
Multidisciplinary case review allows teams to reassess inheritance patterns, evaluate gene–disease relationships, and interpret findings together. This approach can reduce unnecessary repeat testing, limit misclassification of variants of uncertain significance, and support more timely clinical decisions.
Many healthcare programs now formalize this collaboration through regular genomic review meetings, shared data platforms, standardized phenotype documentation, and ongoing reinterpretation of genomic data. These practices can help reduce diagnostic delays and improve the chances of reaching a clinically useful diagnosis.
How do issues such as sample quality, phenotypic data capture, and incomplete clinical information affect diagnostic yield?
RT: Diagnostic yield depends on several factors, including how samples are collected and handled, the quality of the sequencing system, and how comprehensive the genomic test is.
Sample quality is critical. Poor-quality or degraded DNA can reduce sequencing coverage, introduce artifacts, and make it harder to detect certain variants – particularly structural variants, repeat expansions, and changes in complex genomic regions. When DNA quality is compromised, confidence in variant detection decreases, and important variants may be missed during analysis.
Accurate phenotyping is also essential. Interpreting genetic variants depends heavily on how well they match a patient’s clinical presentation. If clinical information is incomplete or poorly documented, variant prioritization becomes less reliable. Using detailed and standardized descriptions of patient symptoms helps improve the chances of identifying the correct genetic cause.
Clinical context further strengthens interpretation. Information such as family history, inheritance patterns, ancestry, previous testing, and treatment response can all help laboratories classify variants more accurately. Without this context, laboratories may take a more cautious approach, resulting in more variants being labeled as “uncertain significance” and fewer definitive diagnoses.
How has next-generation sequencing (NGS) changed the diagnostic landscape for rare diseases over the past decade?
SM: Over the past decade, NGS has significantly advanced rare disease diagnostics. Exome sequencing allowed clinicians to analyze thousands of genes at once, greatly expanding the number of conditions that can be identified compared with targeted gene panels. In some regions, such as the UK, short-read WGS has even been adopted as a first-line test for suspected rare genetic disease.
Despite these advances, more than half of rare disease cases still remain unsolved. Long-read sequencing is now helping address this gap by providing a more complete view of the genome. Its higher accuracy and ability to resolve complex genomic regions – such as paralogous genes and structural variants – can improve the detection of disease-causing variants and provide greater confidence in genetic findings.
In your experience, when is whole-exome or WGS preferable to targeted gene panels?
SM: Targeted gene panels can be useful when a patient’s symptoms clearly indicate a well-characterized condition with a known set of causative genes. However, many rare disease presentations are complex, overlapping, or atypical. In these cases, whole-exome or WGS is often preferred because it allows clinicians to examine the genome more broadly rather than limiting analysis to a predefined gene list.
Long-read WGS can further improve diagnostic yield by providing a more complete view of the genome in a single test. Instead of reconstructing the genome from many short DNA fragments, this approach sequences longer stretches of DNA. This helps preserve genomic context, improves accuracy in complex regions, and enables phasing – allowing scientists to determine which variants are inherited together on the same chromosome, an important factor in diagnosing rare genetic diseases.
What infrastructure is essential to maximize the value of genomic data?
SM: Generating genomic data is only the first step; turning that data into clinical insight requires strong bioinformatics and collaboration. Robust analysis pipelines are needed to accurately detect and prioritize genetic variants, especially complex structural changes and repeat expansions that are often linked to rare diseases.
Population-scale genomic datasets are also important. Databases that reflect global diversity help researchers interpret how common a variant is and identify disease risks that may vary across populations. In addition, secure data-sharing platforms allow researchers to analyze harmonized datasets together, improving variant interpretation and increasing the statistical power of rare disease studies. By enabling access to insights that would be difficult to obtain from isolated datasets, these platforms can accelerate discoveries that ultimately benefit patients.
What emerging technologies are most promising for rare disease diagnostics?
SM: Several emerging technologies are expected to advance rare disease diagnostics. Long-read sequencing is particularly promising because it can generate more complete and accurate genomic data. By reading long stretches of DNA, it can span complex repeat regions and determine their exact length and structure. This improved resolution can strengthen variant interpretation and help identify disease-causing changes that may be missed with other methods.
Artificial intelligence (AI)–assisted variant interpretation is also gaining attention. One major challenge in sequencing workflows is reviewing the large number of genetic variants identified in each genome. AI tools can help prioritize variants, detect patterns, and support interpretation, potentially reducing the time and effort required from clinical teams.
Finally, integrating genomic data with digital pathology and detailed clinical phenotyping may further improve diagnostic accuracy. Combining molecular data with clinical, imaging, and histopathology information can strengthen genotype–phenotype correlations, improve diagnostic confidence, and help clinicians reach answers more quickly.
If you look ahead 10 years, what would an ideal rare disease diagnostic pathway look like?
RT: In the next 10 years, the ideal rare disease diagnostic pathway is likely to become more proactive, automated, and continuous, rather than reactive and fragmented.
AI may play a role in identifying patients who could have a rare disease by analyzing data already present in healthcare systems, such as laboratory results, imaging, clinical notes, growth patterns, medication histories, and evolving symptoms. Early studies suggest AI can help detect rare disease patterns at a level comparable to expert clinicians. In the future, these tools could be integrated into clinical systems to flag potential cases earlier, sometimes even before a formal genetics referral is made.
WGS is also expected to become a first-line test for many patients with suspected genetic conditions. Instead of ordering multiple sequential tests – such as gene panels, exome sequencing, and additional follow-up assays – clinicians may begin with a comprehensive genome test capable of detecting many types of genetic variation in a single analysis. In some health systems, genomic sequencing could even begin at birth.
Interpretation of genomic data may also become more dynamic. Rather than producing static reports, results could be updated over time as new gene–disease relationships are discovered. This is important because many patients remain undiagnosed after initial genomic testing. Routine reanalysis of genomic data could help identify diagnoses that were not previously recognized.
For patients with severe or complex conditions, diagnostic testing may increasingly combine multiple data types. For example, a critically ill infant might undergo rapid WGS alongside other tests, such as pathogen detection or transcriptomic analysis, to provide a more complete picture of the underlying disease.
Finally, diagnosis itself may become only one part of a longer process. Long-term tracking of clinical outcomes, treatment responses, and disease progression could be integrated into care systems, helping clinicians better manage patients and improving future diagnostic approaches.
Overall, the future of rare disease diagnostics may rely on better integration – bringing together AI-supported case identification, genomic testing, ongoing data interpretation, and long-term clinical follow-up to support earlier and more accurate diagnoses.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
