Washington Post | If you put 10,000 people’s genomes in the cloud, could you demystify autism?
“The world,” the narrator notes in Mark Haddon’s novel “The Curious Incident of the Dog in the Night-Time,” “is full of obvious things which nobody by any chance ever observes.” The novel is a mystery. The narrator is a 15-year-old boy named Christopher John Francis Boone. He is autistic, and he is quoting Sherlock Holmes – who was not speaking about autism, in the original Arthur Conan Doyle story, but perhaps could have been.
For all the strides researchers have made in recent years in understanding the disorder we generally think of as autism – actually, it’s a group of disorders, which collectively affect an estimated one out of 68 children in America – so much of it remains mysterious. Scientists are only beginning to untangle which genetic and environmental factors cause the disorders on the autism spectrum.
But there are clues to autism, and lately, many of them have come from studying what amounts to the mapped DNA of people on the spectrum. A major autism non-profit, backed by a tech giant, is betting there are a lot more breakthroughs where those came from.
On Tuesday, the nonprofit, Autism Speaks, will announce the details of an effort to sequence the whole genomes of 10,000 people in families affected by autism, and to share the results with researchers around the world on the Google Cloud Platform. It is sort of like digitizing a DNA library. The effort is called MSSNG; the dropped vowels are meant to show all the things about autism that no one has yet observed, as Holmes might have put it.
“I don’t think there is a database as complete and big as this one anywhere in life sciences,” Rob Ring, the chief science officer at Autism Speaks, said in an interview. The best-case scenario for it, he added, “is that we will provide the data on which discoveries around autism are made for years to come.”
Until very recently, genome sequences were expensive to come by, and thus relatively difficult for researchers to get their hands on. Now, Ring says, the cost is falling to about $2,500 per person. Meanwhile, Autism Speaks has been collecting volunteered sequences for years; it will open its cloud portal to scientists early next year with 1,000 of them already in the bank. Researchers anywhere in the world will have the opportunity to tap in and look for new clues in that data.
The goal is for that access to produce a sort of crowdsourcing for autism answers. David Glazer, the engineering director at Google, compared it to using Google Translate on a Web page, a process that is powered by patterns detected in human translations of words from one language to another. The quality of the translations, he said in an interview, has improved with the volume of those patterns that Google can comb. The same, he suggested, could be true for autism researchers.
“That’s the kind of thing,” Glazer said, “where, if you add enough new information, you can get really big new things.”
Autism Speaks has already amassed 12,000 DNA samples from volunteers across the country. Glazer said Google is enforcing “robust security” to protect the privacy of genomes and the individuals who supplied them.
It’s impossible to know, of course, what scientists will find in the cloud and how their discoveries might help people with autism. “Whether or not this data leads to treatment, we can’t say for sure,” said Matthew Higgins, the chairman of the MSSNG initiative, and the father of a 7-year-old girl, Collette, who is on the autism spectrum. “But it’s certainly going to lead to answers.”
Ring, the Autism Speaks chief scientist, said the group is already thinking about growing its shared stockpile of genomes from 10,000 to 100,000. The more sequences, hopefully, the more answers. As Christopher John Francis Boone observes, in his own detective story, “Lots of things are mysteries. But that doesn’t mean there isn’t an answer to them. It’s just that scientists haven’t found the answer yet.”