Rare Disease Genomics
From Diagnostic Odyssey to Molecular Diagnosis

A rare disease is defined by its prevalence, not its severity. In the US a disease is rare if it affects fewer than 200,000 people (roughly 1 in 1,500). The EU uses a stricter threshold of fewer than 5 in 10,000 (roughly 1 in 2,000). Japan applies an even narrower 4 in 10,000.

These regulatory definitions matter because they determine which diseases qualify for orphan drug designation, the status that incentivizes pharmaceutical companies to develop treatments for conditions that would otherwise not be commercially viable.

Shared characteristics

• Almost always chronic and often progressive
• Most have onset in childhood, though many present in adulthood
• Frequently affect multiple organ systems simultaneously, creating complex overlapping presentations
• Most lack approved treatments, or have only partially effective ones
• Genetically heterogeneous: the same clinical presentation can result from variants in dozens of different genes

How many are genetic?

Approximately 80% of rare diseases have a genetic cause or significant genetic component. Most are Mendelian disorders, caused by pathogenic variants in a single gene following defined inheritance patterns. The remaining 20% include rare infectious diseases, rare autoimmune conditions, rare cancers, and conditions with multifactorial or unknown etiologies. This 80% figure is why genomic sequencing has become the central diagnostic tool in rare disease: for the vast majority of patients, the answer, if it exists, is written in the DNA.

The diagnostic odyssey is not a metaphor. It is a documented clinical phenomenon with measurable dimensions. On average, a patient with a rare disease waits 5 to 7 years from symptom onset to a confirmed molecular diagnosis. During that time, they typically see 8 or more physicians, receive 2 to 3 misdiagnoses, and undergo a cascade of tests aimed at conditions that turn out not to explain their symptoms.

Genomic sequencing does not approach diagnosis hypothetically. It does not ask "could this be condition X?" and test for X. It sequences the entire exome or genome and asks: what is actually there?

For rare disease, this matters for three reasons:

It bypasses the requirement for a correct hypothesis

A patient with an atypical presentation of an ultra-rare condition does not need a physician who has seen that condition before. The sequencer finds the variant. The analytical platform identifies the gene-disease relationship.

It compresses time

A molecular diagnosis that might have taken 5 years of sequential testing can be achieved in weeks with exome or genome sequencing. Published studies have documented 40 to 60% reductions in time to diagnosis when genomic sequencing is used as an early-line test.

It opens doors clinical diagnosis cannot

A molecular diagnosis enables access to disease-specific treatments, natural history studies, patient registries, and clinical trials that are inaccessible without a confirmed genetic etiology. For many rare disease patients, naming the condition is the prerequisite for everything else.

Not all genomic tests are equivalent. Choosing the right test for a specific patient requires understanding what each approach can and cannot detect. The algorithm has evolved substantially over the past decade and continues to shift as costs fall and evidence accumulates.

The trend across clinical genetics programs globally is toward earlier use of WES and WGS, using comprehensive sequencing as a first-line rather than last-resort test, because the cost-per-diagnosis is lower when comprehensive testing is performed upfront rather than after years of inconclusive panel testing. See the WES vs WGS guide for the cost-per-diagnosis argument in detail.

For patients and families living through the diagnostic odyssey, the question that matters most is not which sequencing technology was used. It is: what changes when we finally have an answer? The answer is: more than most people outside clinical genetics appreciate.

Sequencing technology is no longer the limiting factor in rare disease diagnosis. Modern sequencers can produce a complete exome in hours and a complete genome in a day. The bottleneck is what happens to the data afterward.

A WES run produces ~25,000–40,000 variant calls per patient. A WGS run produces ~4–5 million. The clinical team must reduce that number to a handful of candidates, evaluate each against the patient's phenotype and published evidence, apply ACMG classification criteria, and reach a defensible clinical conclusion, often within days, under pressure from families who have waited years.

The VUS problem

Variants of uncertain significance are the most common result of diagnostic sequencing for rare disease patients. VUS rates of 30 to 50% are typical for unselected populations undergoing WES. A VUS is not the end of the road. It is the beginning of an evidence-gathering process. Over time, as more patients with the same variant are identified, as functional studies are published, and as clinical databases accumulate more data, variants are reclassified. Labs need processes for tracking VUSes and proactively notifying families when reclassification occurs.

Reanalysis: a diagnosis hidden in existing data

One of the most underappreciated tools in rare disease diagnostics is re-analysis of existing sequencing data. Because knowledge of gene-disease relationships expands rapidly, a WES or WGS dataset that was non-diagnostic at the time of testing may contain the answer today.

Studies consistently show that re-analysis of WES data yields new diagnoses in 10 to 20% of previously negative cases when performed 2 to 3 years after initial analysis. The variants were in the data all along. The knowledge to interpret them had not yet been established.

This creates a clinical obligation. Labs and clinical programs should have systematic processes for identifying patients who may benefit from re-analysis and initiating it proactively, not waiting for families to request it. VSWarehouse enables exactly this: tracking variant classifications across a lab's full patient cohort and alerting clinicians when external databases reclassify variants previously observed in their patients.

What good interpretation infrastructure looks like

When a rare disease diagnosis is established in one family member, the genomic work has only begun.

Cascade testing

Targeted testing of biological relatives for the specific variant identified in the affected family member. Fast (typically a few days turnaround), inexpensive, and clinically actionable. For autosomal dominant conditions, first-degree relatives each have a 50% prior probability of carrying the variant. For autosomal recessive conditions, carrier testing of siblings and extended family informs reproductive planning. For X-linked conditions, maternal carrier testing is the usual starting point.

Carrier screening

Testing individuals without symptoms to determine whether they carry a single copy of a recessive disease variant. Most clinically impactful when performed preconceptionally: for families with a known recessive rare disease variant, preconception carrier testing of partners and reproductive decision-making can prevent disease in future generations.

Prenatal and preimplantation testing

For families who carry pathogenic variants in known disease genes, prenatal diagnosis (amniocentesis, CVS) and preimplantation genetic testing (PGT in the context of IVF) are available options. Both require a known familial variant, another reason establishing a molecular diagnosis in the affected family member is the critical first step.