When whole genome sequencing identifies a change in a gene that is already well understood and clearly matches a person’s symptoms, the finding is highly reliable. Across a recent meta-analysis pooling multiple clinical studies, whole genome sequencing identified a genetic cause in roughly 3-in-10 previously undiagnosed individuals overall, compared with a little over 2-in-10 for whole exome sequencing; this overall difference was described as a modest advantage that did not reach statistical significance, meaning it could reflect chance in some of the pooled studies. The advantage was clearer, however, within specific clinical groups: among critically ill patients, whole genome sequencing identified a cause in close to 4-in-10 cases compared with roughly 3-in-10 for whole exome sequencing; among those with neurologic conditions, the gap widened further, to nearly 6-in-10 compared with 4-in-10; and among those with multiple congenital anomalies, roughly 5-in-10 compared with under 4-in-10. Among people who did receive a positive diagnosis, whole genome sequencing and whole exome sequencing were similarly likely, at roughly 6-in-10 and 5-in-10 respectively, to change medical management in some way. Families should understand, even so, that more than half of those tested with either approach will not receive a definitive genetic answer on a first round of testing.

A second important limitation, shared with whole exome sequencing, is the variant of uncertain significance: a genuine, confirmed genetic change for which there is not yet enough scientific evidence to say whether it causes disease or is harmless variation. Because whole genome sequencing reads far more of the genetic code than any other clinical test, this kind of ambiguous finding is common, and a geneticist will explain clearly when a reported finding falls into this uncertain category rather than being a confirmed cause; classifications are regularly revisited as scientific understanding grows.

Whole genome sequencing can also uncover secondary findings: genuine, medically significant findings unrelated to the original reason for testing, such as a gene linked to a treatable heart condition or a hereditary cancer risk. The American College of Medical Genetics and Genomics maintains a specific list, currently covering 84 genes where early knowledge is considered to change medical management, and families are always given the choice, discussed during pre-test counseling, of whether they wish to receive this kind of unrelated information at all.

Finally, whole genome sequencing has its own technical blind spots. Short-read sequencing, the technology used by most clinical laboratories today, still struggles with certain highly repetitive stretches of the genome, cannot reliably detect very low-level mosaic changes present in only a small fraction of a person’s cells, and does not capture chemical modifications to DNA that affect gene activity without changing the underlying sequence. Newer long-read sequencing technologies are beginning to address some of these gaps but are not yet standard in most clinical settings.