Michael Schatz: On Autism and De Novo Mutations | Big Think

So in autism, I collaborate with some folks at Cold Spring Harbor Lab where we're participating in a project that's been sponsored by the Simons Foundation.

And the idea there is over the last several decades, different hospitals around the U.S. have been identifying families where one kid will have autism, their sibling does not, and have been collecting from those people blood samples of the autistic child, their sibling, and also their parents.

So then the questions tend to be: well, what is it about these autistic kids in terms of their genome that are special or changed or different, relative to their parents, relative to their siblings, that either predisposed or perhaps even caused the disorder?

There's been a lot of interest. There's been a lot of study about this. There's a lot of uncertainty about this.

And really, it's been the growth of DNA sequencing technologies that has made it possible to finally drill down into the genome to see what exactly is special about kids with autism.

So through some early studies, there was a lot of evidence that so-called de novo mutations were at least partially responsible for autism.

So a de novo mutation is: most of your genome is a combination of your parents' genome, so it's half your mom plus half your dad put together forms you.

But a very, very small number of changes are so-called de novo, meaning that there'll be spontaneous changes in the genome.

There had been some earlier work through Mike Wigler's Labs and others that saw that there were large-scale of what's called copy number changes that were associated with autism, where segments of the genome would either be duplicated or segments of the genome would be deleted and lost in kids with autism.

Now through improved sequencing technology, we can drill down and we can look and see: well, what are the specific bases that are being deleted or amplified? What are the specific mutations?

It just gives us incredible power to be able to identify these changes.

We're still relatively early in the project.

So far, we've sequenced about 300 such families. The full project will be about 3,000 families, and the hope there is to be able to identify the patterns that lead to the disease.