If genome sequencing examines more DNA than exome sequencing, why would anyone choose an exome? And after a negative exome, is a genome always the next step? The answer depends on the clinical question and on what the laboratory actually analyzes, validates and reports.
A useful comparison goes beyond the name of the technology. Ask what information is missing and which service is most likely to supply it.
The basic difference—and the important qualification
Exome sequencing focuses mainly on protein-coding regions. Genome sequencing samples DNA much more broadly, including regions outside those coding segments. MedlinePlus provides an introduction to both approaches. Broader data collection does not mean that every region is measured equally well or that every observed difference can be interpreted.
A laboratory may collect genome-wide data but initially analyze a defined gene set. Another may include several kinds of structural variation or repeat analysis. The GeneReviews overview of testing approaches describes why analysis scope and laboratory capabilities matter.
| Question | Information to request |
|---|---|
| What is sequenced? | The targeted regions, specimen and important coverage limitations. |
| What is analyzed? | The genes and variant classes included in the validated analysis. |
| What is reported? | Policies for uncertain findings, additional findings and clinical confirmation. |
| What happens later? | Data retention, reanalysis options, turnaround time and costs. |
When do professional guidelines support broad sequencing?
The ACMG’s 2021 evidence-based guideline recommends considering exome or genome sequencing as a first- or second-tier test for children with congenital anomalies beginning before age one, or developmental delay or intellectual disability beginning before age eighteen. Its scope matters: it is not a claim that either method is the best first test for every genetic concern.
A clinician can explain whether the presenting features call for broad sequencing, a targeted assay or a combination. Previous results and the urgency of the clinical question belong in that discussion.
Example 1: the old data may contain a newly interpretable answer
Fictional teaching example. A child’s exome was nondiagnostic several years ago. A gene now has convincing evidence linking it to a condition that fits the child’s evolving features. The original data adequately measured the relevant region.
Reanalysis may be worth discussing because the missing ingredient could be interpretation rather than new sequencing. That possibility is not a guarantee: the laboratory must check the old data’s quality and the current evidence.
The ACMG’s reanalysis guidance distinguishes reviewing existing results from retesting with another method. This statement was reaffirmed in 2024. Ask what new data or knowledge each proposed option would add.
Example 2: the suspected mechanism is outside the earlier analysis
Fictional teaching example. A person has a negative exome report, but subsequent assessment raises a specific repeat-expansion disorder. The report does not include validated testing of that repeat.
The useful question is whether the next assay can assess that particular expansion, including any relevant sizing or other molecular features. Ordering a test called “genome sequencing” without checking its validated repeat analysis does not answer the question.
For a real biological example, GeneReviews on FMR1 disorders explains the importance of repeat-expansion testing and, where appropriate, methylation assessment. This illustrates why the mechanism should guide the method; it does not imply that everyone with a negative exome needs FMR1 testing.
Example 3: a difficult region needs a performance question
Fictional teaching example. A clinician suspects a condition involving a region with highly similar DNA elsewhere in the genome. One report lists the gene but also identifies a relevant coverage or mapping limitation.
The question for the next laboratory is specific: can its method distinguish the relevant DNA and detect the suspected type of variant there? More sequence data are useful only if they address the actual gap. A validated targeted method or another sequencing strategy may be appropriate, depending on the case.
The ACMG technical standard for next-generation sequencing discusses laboratory performance and validation. Our pseudogene-interference guide explains why similar sequences can be challenging.
Read diagnostic-yield percentages with their denominators
In Wojcik and colleagues’ 2024 study, genome-based investigation produced molecular diagnoses in 218 of 744 previously undiagnosed families in the initial cohort, or 29.3%. In 61 families—8.2% of that cohort—the authors judged that genome sequencing was required to identify the relevant variants.
These numbers answer different questions. The overall yield after a genome investigation is not the same as the fraction requiring genome sequencing. Some answers could be recovered through updated exome analysis or additional analysis of existing exome data. This was a selected rare-disease cohort, not a personal success forecast for everyone considering testing.
When comparing studies, write down the prior workup, inclusion criteria, family samples and definition of a diagnosis. Then ask whether the reported advantage arose from the measurement, the analysis or knowledge that became available later.
A practical route through the decision
- State the unresolved question. Describe the clinical concern and what previous testing did or did not establish.
- Identify the likely gap. Is it an unmeasured region, a variant class, sample choice or interpretation?
- Compare realistic options. Consider reanalysis, targeted testing and broader sequencing with the clinical team.
- Check the proposed service. Ask about validation, family samples, timing, costs and consent.
- Plan for an uncertain or negative result. Agree on who explains it and how follow-up is arranged.
Our guides to trio sequencing, secondary findings and testing limitations cover related decisions. Use the paper-comparison workspace to keep study claims separate from the capabilities promised by an actual laboratory.