Jay Shendure's lab at the University of Washington has shown that exome sequencing can be used to find a gene behind a rare genetic disorder. Using an array- and next-generation sequencing-based approach, Shendure and his colleagues sequenced the exomes of 12 people: eight individuals from the HapMap Project and four people with Freeman-Sheldon Syndrome, an autosomal dominant disorder marked by craniofacial, hand, and foot abnormalities.
"For us, the study really had two goals. One was to evaluate how one goes about doing targeted resequencing with next-generation technologies and, within that larger problem, there's the specific problem of the exome," says Shendure, who led the Nature study. "And the second goal was to try to put this into a disease context, rather than just being a proof-of-concept around technology itself."
To home in on the exome, Shendure and his colleagues used array capture followed by next-generation sequencing on an Illumina machine. Instead of working to optimize that initial capture, the researchers used a second array to find data missing from the first. They yielded an average of 6.4 gigabases per person.
Then they turned to looking for MYH3, the gene behind Freeman-Sheldon, to show the method's applicability to identifying Mendelian diseases. Shendure says they chose to study this disease because the answer was already known. "We sat down and talked through what might be a challenging disease — in the sense that it's autosomal dominant, so it's going to be more challenging than a recessive [disease]," he says. "And we had multiple individuals, so we could set up the equivalent of what we felt was a real-world situation."
For that search, Shendure and his colleagues catalogued variants from the exomes of the people with Freeman-Sheldon, looking for genes with non-synonymous variants or indels. They then filtered out common variants from those 2,000 candidate genes by using dbSNP. That alone, Shendure says, was inadequate, so they also used variations from the HapMap individuals as an extra filter. Together, those narrowed the candidate gene list down to one, MYH3. "[This] presages what is going to happen with 1,000 Genomes and projects like that," Shendure adds.
He and his lab are currently working on applying this method to Mendelian diseases caused by unknown genes, as well as more complex diseases.