Latest / Computer Says Maybe / Short: What Just Happened to 23andMe? w/ Jenny Reardon
Transcript
- Alix: [00:00:00] Welcome to Computer Says maybe this is your host, Alix Dunn. And this episode's a bit of an experiment. We realize that there's a lot of guests we have on the show and people in our network who have really interesting context on newsworthy topics, and maybe we don't wanna record a whole. Episode about it, but it would be nice to be able to allow our listeners to tap into that expertise when newsworthy stuff's happening. We actually came up with the idea when hearing about 23 and MES bankruptcy proceedings, and having just had Jenny Reardon on the show who is a historian and expert on the way that genomics and the field of genomics has developed over the last 30 years. With lots of impacts on issues of equity and inequality and injustice and impacts on marginalized groups. We had an amazing conversation. Highly recommend you checking it out if you haven't heard it. But given her expertise and the case with 23 and me, we [00:01:00] figured we'd have her on the show just to give a quick kind of run through of what she thinks about what's going on and putting it into context of companies that have come before. Jenny: Well, I'm Jenny Rearden and I am the founding director of the Science and Justice Research Center at uc, Santa Cruz, and a professor of sociology there and have written a couple of books on genomics and impressions of genomics and race. The first one, race to the Finish, identity and governance in the Age of genomics, and then the post genomic condition, ethics justice knowledge Alix: after the genome. So basically 23 and Me says, Hey, send us your. DNA and we'll like tell you interesting things about yourself and then lots and lots and lots of people did that. And then 23 and me couldn't find a business model I guess. 'cause once you do this once for someone, they're not gonna come back 'cause they don't have changing genetic material. Um, they ran outta money. Then they went bankrupt. And then when you're going through bankruptcy, [00:02:00] you get asked, what can you sell to pay back all the people that you owe money to? And an obvious offer is all this information about all these people. How much do you know about what's going on with the case? Have you been anticipating this? What do you think about it? Jenny: It's funny, when I was looking back, what I wrote about this in my book, the Post Genomic Condition book, which was published. End of 2017. I have a chapter in there on 23 and me, and I was just reviewing that in which I say there's a high likelihood that 23 and me would follow the same course that Navigenics followed, which was one of the, at the time of the writing of that book was one of the three I. Direct to consumer genetic testing. Companies that were there that was in existence, decode me, had already gone under, Navigenics, went under in 2015. It wasn't a big leap to think that maybe 23 and me would be following the same. Path except for the fact that 23 and me had a lot more, you know, it was based in Silicon Valley. It had [00:03:00] always had a lot stronger backing. And Wiki was married to Sergio, Bryn, who was co-founder of Google. Um, oh, wait, Alix: wait, wait, wait. The founder of 23 and me was married to Oh yeah. The Google found. I didn't know that. Jenny: Yeah, yeah, yeah. When I was doing interviews at the time, people I would talk to at the company would make a big point about how we're not Google by any stretch of the imagination, but of course they were literally in bed with folks at Google. So, um, and, and the whole idea was actually that Linda Avi. Who was the co-founder, they were studying Parkinson's. There is a history of Parkinson's in the Brynn family, and so Sergio Brynn was very interested in this and went to a meeting and started asking technical questions like, what's the P value on your data? AV was saying we don't have high confidence, statistical confidence 'cause we only have 300 Alix: samples. So they were like, how can we increase the sample [00:04:00] size of genetic material that we can test against? Let's build a direct to consumer genetics Jenny: company. Yeah. Yeah. There's a whole history it very interesting genealogy here that weds a kind of. Techno utopianism of a certain kind. You know, California techno utopianism powered the people through informatics with corporate, you know, corporate. I mean, that is the California brand, I guess. So 23 and Me is just part of that. And when 23 and Me comes along though, it has a lot more, you know, the idea was Google your genome, marry the power of this massive data management system. Google with the power of genomics, which is big data par excellence. So, so that, that made sense. And we're gonna wet it with this idea of democratization, and we're gonna argue that the federal government is paternalistic. You should not trust the state with your genome. You should [00:05:00] trust us. We're democratizing this. And you can imagine the amount of trouble they face because they were really trying to bring the whole, we're Democrat, you know, the internet, the ethos of the internet we're democratizing informatic systems. But now we're bringing it to the genome, which is just information, but it's not just information, it's also your bodily materials. It's about biology, which is a. Very complicated domain. It's about your health and, and it wasn't clear that people wanted to share all their health data because of all the reasons why you wouldn't want us to share your health data. So there were a number of challenges that the company faced from the very, very beginning, but it was backed by a lot of powerful stuff. You know, it was backed by. The Google space, you know, however, even if they, they weren't the same company, they were in those networks and they were able to fundraise in those networks and they used those networks quite effectively. The first people who ever Googled their genomes, you know, [00:06:00] or, or spit in the tube for 23 and me. We're like the Digerati and they were the rich and famous Fashion Week in, in New York City in 2008. I think there was a party, a spit party, you know, and there were famous people doing this. What makes this story, I think interesting is, I mean, first of all, 23 and Me is the longest survivor. It did manage to collect 15 million samples, which is, makes it, I think, the largest DNA database. I think in the world, I could be wrong. So they were successful at collecting more data than anybody else. You know, the, the federal government has. The goal of collecting a million people's genomes through all of me, and so you could say they're 15 times bigger than the federal government's database. What was it really unique about? What 23 and Me was doing was not the genotypic data, it was the phenotypic data. It was the relationship they were building with the customers, a trusted [00:07:00] relationship with their customers to get those customers over time to share more and more data. Because if you're going to make any of this genotypic data of any use, you have to correlate it with something. It has to become meaningful. And the whole problem that I actually talk about in the post genomic condition was that the problem of genomics is the problem of meaning. You can sequence all that DNA. But what the heck does it mean? And that has been the real problem at the heart of genomics, that since the sequence and even before the sequence was done, people have been trying to solve. But unfortunately, the apparatus that genomics focused on the whole time was, well this figure out the sequencing and we're gonna get really, really good at sequencing and we'll get to the meaning part later. So the Human Genome project was just as the technology to do the sequencing. But the sequence was never gonna mean anything if you didn't have people's phenotypic data. So 23 and Me was stepping in to solve that. And if you're gonna get people's [00:08:00] phenotypic data, they need to trust you to give you that valuable information. Alix: What kinds of information do they collect? Like how are they doing it? Jenny: Well, they started out, well, it was through surveys and you know, they're creating a platform. They put a lot of money into making that platform friendly, engaging, and fun. So they did a whole thing where they invested in what colors are we going to use? And if you know, 23 and me has these cool colors and 23 and me had. Swag, like it had beanies and t-shirts and you know it, and it put out fun videos about what is DNA. They were trying to make themselves friendly, you know, because genetics had this aura of being scary. You know, people think eugenics or they think I have a genetic disease and I'm gonna die, so they don't wanna interact with it. So 23 and me put a lot of emphasis on that, and then they put up these surveys. And the surveys to begin with were fun surveys like. Can you smell your pee when you eat Asparagus was one of the questions, you know, Alix: so then in future you could sequence your genome and be like, aha. A fun fact we can [00:09:00] tell you about yourself is it's highly likely that you can smell asparagus when you. Jenny: P. Yeah. Yeah. In fact, they had a little button that had asparagus on it in the shape of chromosomes. It was trying to make people feel this was a fun space and they could be a part of this cutting edge science and technology. They could be part of the digerati, the spit. The other thing is I thought was so interesting is that the first group that they reached out to, one of the first groups they reached out to, 'cause they were aware that actually this was quite risky for people at the time that they started the Genetic Non-discrimination Act did not exist. And so there were risks. There still are risks. In fact, one of the issues with this bankruptcy is whoever inherits this, if they manage to find a buyer, if they do manage to restructure the data, still it's not protected A by hipaa, B by long-term care insurance. You know, you can be discriminated against because you know [00:10:00] the basis of this. So not all forms of. Alix: Health matters are protected. So you could imagine an insurance company buying this up and saying, great. We have genetic data exactly about a pool of customers that we can now discriminate against and set prices or premiums, or deny them coverage altogether because they're in this and they have a particular Right. Okay. Jenny: And, you know, 23 and Me was very careful about this to begin with. They were like, we won't share your data for these purposes. That might hurt you. And, and they do celebrate on their website. Even though we've had requests from law enforcement, we have never shared your data. I trust 'em. They haven't shared their data. So they have tried very hard to keep the trust of their 15 million customers, but whoever they sell it to, you know, could change the privacy agreement. They could agree to it and then change it. And this was always the trouble And, and whenever I taught classes about this and I continue to teach classes, I always did my. Public information messaging to my students, because often they just were like, oh, 23. [00:11:00] And me, you know, my aunt or my uncle, or my mother and my father. Like, let's find out where we're from. This'll be fun. And I've always said, okay, that might be fun, but you really need to think through, do you really want this company to have all of your DNA. Data because it could be used in these other ways. Like you could be denying long-term care based on this. They say they won't share with the government, but the government can subpoena it under, you know, could get access to it. And you know that's all fine and good when you trust the government and the government is doing things you believe in. But as we know, the government changes. It's actually an interesting moment to be thinking about this because it raises a question, who would you trust to share your DNA with? Would you trust 23 and me? Would you trust the NIH? And it used to be that that distinction you might say, okay, I trust the public more. I was just listening to the new NIH head talk about their vision. And the vision is all about [00:12:00] creating these data platforms that link all the data together, including all the genomic data that the government has with all the data. They have across health and human services, and who knows what they're gonna do with that data and for what ends? And I'm not sure, I would say if you asked me, who would you trust more 23 and me or the government to share your data with, I'm not sure what I would say right now. I think the wrong story about 23 and Me is if you wanna boil it down to, oh, it's wrong for corporations to take your data. I think this kind of corporate takeover of science. Yes. Okay. There's things that I think we should think about in terms of the specificity of corporations. They operate in much shorter funding cycles. They don't maybe have guaranteed five years of funding from the government, so they are being driven by short term profit goals. Okay? So that might be a difference and that might be a reason why you don't wanna share your data with corporations. But whether or not it's a corporation or whether or not [00:13:00] it's a government, there are very similar questions of trust. Who's deciding how this data is gonna be used? Do you trust them? And the other thing is that both the public effort and these corporate efforts face the problem that it's really hard to make anything meaningful out of genomic data. And that's part of going back to your point, yeah, they didn't have a winning business model because there's only so much you could try to get people to engage with the website once they've got that, you know, where are we from in the world out of the way. The other part about creating health information is incredibly difficult. And that's true across the board. And I also just think it's interesting to think about all that data. Like if you participated in a public study, an NIH funded study, and you gave your sample and you gave all your data, you're presuming that the federal government exists and continues to exist and there continues to be an NIH. But given the current moment, it's possible to think about a moment when that's not the case. So then what happens when the NIH goes under, you know, it's really, yeah, Alix: or they try and privatize. Certain parts of it. [00:14:00] Jenny: Yeah. It's very likely that we may be seeing some of that. So in fact, I need to continue listening to the current head of N nih and they're talking about public-private Alix: partnerships. I knew that you were gonna have the dirt on this and I also knew that it wasn't as simple as company going bankrupt with interesting data. Isn't that bad? Uh, there's Jenny: so much. It's actually a super, you, you asked me to do this. It's like, I have tons of things I could say about this. It's super interesting. Alix: Thanks for listening to our first short. If there are topics you see in the news and you're like, Hey, I really wish I could get computer, says maybe's, take on this. Let us know and we can see if there's anybody in our network that can speak to that issue. It also just gives us a better sense of what you would find interesting and also like what news you're paying attention to. 'cause there's so much going on with technology, politics and that would be helpful. Thanks to Georgia Iacovou for helping figure out how to do this, uh, and to Sarah Myles as always for producing the episode.