Big Data, Little Boy

By Wantapeanut @wantapeanut
According to the CDC, 1 in 68 children has been identified with autism spectrum disorder. The rate is five times higher for boys than for girls. These are staggering numbers. And like many parents of autistic children, I'd like to know why the numbers are so high.

To clarify, I'm not looking to blame anyone. I'm not trying to point to a government conspiracy or looking to sue "Big Pharma."

I'm also not looking to cure anyone who doesn't want to be cured. Even the term "cure" is controversial, as autism is not a disease. And, just as many in the deaf community choose not to become hearing, even if that technology is available to them, many autistics do not want to change their neurology. Being autistic is a fundamental piece of who they are, and feel no desire to change. I can respect that.

However, Moe is severely disabled and if there is information that can help me find ways to help him thrive, I want to do that. And I believe at least some of the answers come with information. Lots and lots of information.

This summer, Google and Autism Speaks announced a partnership "to accelerate advances in autism diagnosis, subtyping and treatment as part of Autism Speaks 10,000 Genomes Program." Autism Speaks would like to sequence 10,000 genomes, and will host the data on the Google Cloud Platform.

Like all data, this genome information is morally neutral. What we do with it may not be. Some are concerned that it will eventually be used "cure" autistics, or to detect autism as part of prenatal screening (much like Downs Syndrome can be detected). But I have high hopes for this kind of research. There is so much we don't yet know about autism. Why do some kids respond to early intervention, seeming to "recover" almost completely, while others do not? Why do some kids seem to regress, as Moe did? What are factors leading to relative severity of the disorder, and why are certain co-morbid conditions, like epilepsy, so prevalent? Why might some seem to respond better to dietary or biomedical interventions while others do not?

Today, parents need to try out a host of expensive and time consuming therapies to determine which ones are the most effective for their kids. And with so many options, it is hard to try everything. It would be valuable to know, a priori, what kinds of therapies Moe might respond to best, or where to focus our efforts. 

We must not, however, forget that, our ultimate concern is at the most individual level. That the goal is help autistic individuals and their families. Whatever Big Data may someday tell me about what caused Moe's autism, or what his subtype, he is a human being with his own likes and dislikes. And we humans have a funny way of not always doing or responding the way the data says we should. We must be careful to remember that while their genomes can be classified, autistic people are each unique individuals.

I am a fan of big data. But I'm an even bigger fan of one little boy.

Moe, perusing a toy catalog while
snuggling his Android.

This post was inspired by Dataclysm: Who We Are (When We Think No One’s Looking) by OKCupid co-founder Christian Rudder. Join From Left to Write on October 9th as we discuss Dataclysm. As a member, I received a copy of the book for review purposes.
Visit Autism Speaks to learn more about the AUT10K project or read this story from USA Today.