Soapbox Science

23 and Me CEO Wojcicki on turning 150,000 DNA tests into a research database

23andMe, which started out in 2006 as a “personal genome service,” offers DNA analysis to consumers for $299 over the Internet. Now the California-based company is building a research program with crowdsourced data from its 150,000 customers. Company scientists have published several papers.

Anne Wojcicki

They are working with the FDA to establish standards for consumer-level DNA analysis. Genentech has tapped 23andMe for postmarking research. And this summer, the company bought CureTogether – a website that collects data from online patient networks. Last week, 23andMe co-founder and CEO Anne Wojcicki came to Cambridge, Mass. at the invitation of the local chapter of the Institute of Electrical and Electronics Engineers. She spoke to our Boston blogger, Tinker Ready, after a talk at the Broad Institute auditorium.

Q. Isn’t  self-reported, crowd-sourced  data unreliable?   

That actually hasn’t been as much of an issue anymore because we’ve published so much on our data set…We find the customer is not going to be able to answer questions on every aspect of their medical care. But 90 percent of it, they’ll know. What we really specialize in is the right way to ask a question.  Sometimes you can ask question in a different ways and get slightly different reposes. So we make sure we ask question the right way.

Q. You’ve begun working with pharmaceutical companies on phase 4 studies.  Is 23andMe moving in that direction and away from consumer DNA analysis?

The heart of the business is consumers signing up — and making that a sustainable business. That said, we definitely want to connect more and more with pharma partners to leverage the platform we‘ve created.  The example I gave — phase 4 studies,  following patients over time, understanding how they are using therapies, collecting quality of life data  — it’s something that we’re perfectly set up to do.

Q. How does the genetic component of your work fit into post-marketing studies?

My hope it to avoid another example like Vioxx. We get customers all the time saying they wish it were back on the market. In the genetic world, we would have Vioxx on the market but there would be a label that says “people with this genotype should not take the drug”. You would be able to keep the drug on the market but you would be able say that this group of individuals should not take the drug because it makes them at higher risk for heart problems.

My goals for having the Phase 4 studies with a genetic component is to potentially start to identify either subtypes who are really going to respond to the drug , or subtypes who will have adverse events.

Q. The ENCODE study found that much of what was thought to be useless DNA is actually biochemically active. How does this impact on your work?  

The SNPs (the company analyzes) are across the entire genome so it’s not just SNPs in genes. But it should definitely be very value for our research and the types of studies we’re doing — also, for us to empower other researcher to… understand which findings are meaningful and which ones are not.

Q. You said were bombarded with tough questions when you gave a talk at Harvard last year. Can you describe that event?  

Being in Silicon Valley and being quite used to the Facebook culture of sharing–there was a lot of concern about people sharing. It’s dangerous, it’s an untested world. There was a lot of concern about privacy … I understand all of it. We work a lot with privacy experts and we take it very seriously. To be honest, I loved it. I loved getting that feedback … That’s the only way we’re going to improve and grow.

Q. Did you also say they also raised concerns about the ability of consumers to interpret the data?

With some of those concerns about—can you handle all this information? — there are a number academic groups independent of us who have done studies to show that people can understand this (genetic) information and make sense of it. So, this is becoming less and less of a concern. But people always worry – you’re returning lots of data (to consumers) you don’t’ know what it’s all means? How are you going to handle that?

That’s where we really see ourselves partnering with the academic world and with physicians. Let’s do it in the responsible way. And we’re always open to change.

Q. Last thoughts?  

I would encourage every researcher to question– as you are doing studies – are you engaging they our study participant in this research?  I think there is a huge missed opportunity in treating them as a human subject versus treating them as a true partners in research.

If you want to find out more about 23andMe, this month’s Science Online NYC (SoNYC) topic for discussion was the $1,000 genome – are we ready? We heard from Virginia Hughes, a freelance author who has written about her experience with the 23andMe. You can check out this month’s livestream, or take a look at our Storify collating the online conversation.


Comments are closed.