- Title: Assistant Professor of Genetics, Cold Spring Harbor Laboratory
- Education: PhD, University of Idaho, 2002
- Recommended by: Bruce Stillman
When Jonathan Sebat and his postdoc advisor made the discovery that would bring copy number variation into the limelight, they didn’t fully realize what they were seeing. In fact, Sebat says, he thought something was horribly wrong. After all, the experiment aimed to get a baseline of structural variation by looking at the normal, healthy people who usually serve as controls. “We surfed across the genome with an 85,000 probe array [and found] many different spots where there was clearly a gain or loss of DNA,” Sebat recalls. “The initial reaction was, ‘Oh my gosh, what are we going to tell these healthy people?’”
It’s not that Sebat and his advisor, Cold Spring Harbor’s Mike Wigler, discovered copy number variation. Structural variation and CNVs in particular were known in the genomics community, but were “thought to be an oddity of the genome,” Sebat says. Finding that even normal people have such tremendous variation was the real kicker. “It was really the discovery that this was a ubiquitous property of the genome that was the fundamental thing that changed our understanding and really influenced the field,” he says. Disorders long thought to be simple Mendelian diseases — color blindness and Charcot-Marie-Tooth syndrome, for example — turned out to stem from copy number variation.
Since then, Sebat, who began his career in bacterial genomics, has devoted his work to studying CNVs and other kinds of variation. His lab’s focus is on gene discovery in psychiatric diseases — so far, autism and schizophrenia — and posits that these complex disorders could stem from high rates of structural mutation. A CNV scan in autism patients and their families, for instance, showed a high incidence of spontaneous microdeletions that could account for 10 percent or more of autism cases, Sebat says.
Scanning for CNVs with arrays or other tools gives a big advantage over other approaches, such as typical association studies, Sebat says — namely, that when you implicate a gene or locus in the disorder, “you’re not left wondering where is the causal variant,” he says. “When you find a collection of structural variants, you now have a collection of functional variants."
Up next for Sebat is performing CNV scans with a 2 million probe array, a significant step up from the 85K probe arrays he had been using. “We’re going to be seeing 10 times as much variation with the new platform,” he says.
For the field as a whole, though, Sebat says that the key will lie in data analysis. “We have a lot to learn from the world of SNP association,” he says, noting that so far there aren’t methods to treat structural variation data “so that you can have confidence in your results.” That kind of rigor will be critical to keeping CNVs in good standing with scientists. “If this doesn’t get solved, ultimately we’re just going to get a flood of garbage into the literature,” he says.
Publications of note
Earlier this year, Sebat was lead author on a paper linking autism to structural variation. “Strong association of de novo copy number mutations with autism” appeared in Science and represented the first results of the autism project, says Sebat. “The next phase is going to focus on a larger sample of sporadic families using a much higher-resolution platform.”
And the Nobel goes to…
If Sebat were to win the Nobel, he hopes it would be for work “to uncover the key genes in schizophrenia or autism that would ultimately improve treatment of the disorder,” he says. “I think that there is an autism pathway, and even amongst a diverse set of genes that there will be processes that we can target therapeutically.”