Sequencing of the exome – the protein-encoding parts of all the genes – is beginning to dominate the genetics journals as well as headlines, thanks to its ability to diagnose the formerly undiagnosable.
Pulitzer Prize in Explanatory Reporting honored the Milwaukee-Wisconsin Journal Sentinel’s coverage of a 4-year-old whose intestinal disorder was finally diagnosed after sequencing his exome. Once investigators assigned a gene to his symptoms, a bone marrow transplant saved his life. And a just-published study compared the exomes of 12 children with combinations of developmental delay, intellectual disability, and birth defects at the Duke University genetics clinic to reference exomes, revealing 7 mutations, 2 in genes not known to be associated with disease.
In the best-case scenario, mutations revealed by exome sequencing suggest a treatment, as it did for the 4-year-old. But that may be unusual. "We can’t treat most of the Mendelian diseases we know about, so we won’t be able in the near and medium term to treat most of the cases that are diagnosed by sequencing," says David Goldstein, PhD, director of the center for human genome variation at Duke and an author of the study. The new National Center for Advancing Translational Sciences may add to existing treatments and new drug discovery by providing access to compounds from three major pharmaceutical companies. One study’s reject could be another’s cure.
But for certain types of genetic disorders, exome sequencing won’t help. Understanding what, exactly, an exome is reveals why.
A little less than 2% of the 3.2 billion bases of a human genome encode protein. Most genes consist of sections that are transcribed (into RNA) and translated into protein -- these are exons – and sections that are transcribed but are then snipped out before the protein forms – these are introns. The exome, including only exons, is to the genome what a Wikipedia entry about a book is to the actual book. It’s part of the story, albeit an important part.
Admission: Back in the Precambrian period when I was in high school, I read the CliffsNotes version of John Steinbeck’s "The Grapes of Wrath." I read the actual book many years later, and what a difference! The meager plot summary I read in high school missed the nuances, the connections, the feel and the utter devastation of the final scene.
Analyzing an exome to understand a disease is, in some cases, like reading the CliffsNotes version of a classic book.
The 10 Exceptions
Understanding the limitations of exome sequencing is important because it’s already here. “Be one of the first to get your personal exome sequence,” proclaims 23andMe, about its pilot Exome80x project, offered direct-to-consumer, “for research and educational use only.”
The first CLIA-certified test, Clinical Diagnostic ExomeTM, became available from Ambry Genetics earlier this year. A news release announcing the diagnosis of three tough cases calls the technology "essentially a human genome project for an individual patient." Said CEO Charles Dunlop, "Some of these families have been trying to figure out what was ailing their children for years, and we solved the riddle in weeks."
But exome sequencing won’t help every family, and here’s my list of reasons why. The technology won’t detect:
1. Genes in all exons. A few exons, such as those buried in stretches of repeats out towards the chromosome tips, aren’t part of exome sequencing chips.
2. Mutations in the handful of genes that reside in mitochondria, rather than in the nucleus.
3. “Structural variants,” such as translocations and inversions, that move or flip DNA but don’t alter the base sequence (detectable other ways).
4. Triplet repeat disorders, such as Huntington’s disease and fragile X syndrome. Their mutations don’t change the DNA base sequence – they expand what’s already there.
5. Other copy number variants will remain beneath the radar, for they too don’t change the sequence, but can increase disease risk.
6. Genes in introns. A mutation that jettisons a base in an intron can have dire consequences: inserting intron sequences into the protein, or obliterating the careful stitching together of exons, dropping gene sections. For example, a mutation in the apoE4 gene, associated with Alzheimer’s disease risk, puts part of an intron into the protein.
7. "Uniparental disomy." Two mutations from one parent, rather than one from each, appear the same in an exome screen: the kid has two mutations. But whether mutations come from only mom, only dad, or one from each has different consequences for risk to future siblings. In fact, a case of UPD reported in 1988 led to discovery of the cystic fibrosis gene.
8. Control sequences. Much of the human genome tells the exome what to do, like a gigantic instruction manual for a tiny but vital device. For example, mutations in microRNAs cause cancer by silencing various genes, but the DNA that encodes about half of the 1,000 or so microRNAs is intronic – and therefore not on exome chips.
9. Gene-gene (epistatic) interactions. One gene affecting the expression of another can explain why siblings with the same single-gene disease suffer to a different extent. For example, a child with severe spinal muscular atrophy, in which an abnormal protein shortens axons of motor neurons, may have a brother who also inherits SMA but has a milder case thanks to a variant of a second gene that extends axons. Computational tools will need to sort out networks of interacting genes revealed in exome sequencing.
10. Epigenetic changes. Environmental factors can place shielding methyl groups directly onto DNA, blocking expression of certain genes. Starvation during the “Dutch Hunger Winter” of 1945, for example, is associated with schizophrenia in those who were fetuses at the time, due to methylation of certain genes. Exome sequencing picks up DNA sequences – not gene expression.
3 Great Uses for Exome Sequencing
Exome sequencing is of great value in two obvious situations: (a) finding a mutation in a known gene behind an “atypical presentation,” such as Nicholas Volker, the saved Pulitzer boy; and (b), identifying mutations in novel genes, like 2 of the 7 children in the Duke University clinic.
Another application is subtle: exome sequencing reveals incomplete penetrance, a phenomenon in which a person gets lucky. He or she has mutations that should cause a particular trait or illness, but they don’t.
Exome sequencing of parent-child trios can reveal when an apparently healthy parent actually has the same mutation as the sick child, but for some reason escaped the genetic fate. A genetic counselor would use this information in predicting risk for siblings. If mom or dad contributes a mutation, the next kid faces a much higher risk than if the affected child has a new mutation. But there’s a bigger picture. Figuring out how the parent stays healthy can reveal new drug targets, and perhaps even lead to repurposing an existing treatment.
Happily, exome sequencing has a limited lifetime, because, like climbing a mountain or running a marathon, an end is in sight: knowing what all of our genes do.
This blog first appeared on Scientific American blogs on May 16. Thanks to Marlene Shaw for suggesting the topic.