For my first entry in a series on “blockchain genomics” companies (see previous post for details), I had the pleasure of sitting down with Luna DNA co-founder Dawn Barry. I picked Luna DNA to kick off the series for a couple of reasons. First, they’re right here in San Diego (well, technically Solana Beach), so I had the luxury of conducting the interview in person. Second, the company was founded and being run almost entirely by former colleagues of mine from Illumina – people I really respect and consider friends. Who could pass up that opportunity?!
Shawn Baker (SanDiegOmics): This is such an interesting new space. How would you describe what Luna DNA is trying to do?
Dawn Barry (Luna DNA): At a super high level what we’re doing at Luna is forming a health data community. Our role in that community is to aggregate and organize human health data for the greater good of medical research. Our approach is to break down the centralization of siloed data and putting people at the center. People own their data and their data is valuable. They should have a mechanism to share it in a fair and transparent way and they should be rewarded for sharing it. An analogy would be to think of this almost like a co-op where everybody brings a valuable asset into the ecosystem. As that asset grows and if it has the potential for greater value, that value flows back to the individuals through Luna’s community-owned database.
Shawn: Ok, so many questions! Let’s start with what you mean by “human health data”. What kind of data are you talking about and will Luna be generating any of it?
Dawn: Think of three buckets of the information you have – your genomic information, clinical medical data that’s generated in the healthcare system (like EHR type information), and the third bucket is contextual information – lifestyle, environment, nutrition.
So then your second question was how do you collect that? First, how do you collect DNA files from and from individuals? One source of recruiting members with DNA profiles is looking at individuals who have used a direct to consumer DNA testing product. We estimate there’s about 17,000,000 people that have purchased one of these genetic, at-home test kits and probably have a DNA file sitting idle that they can put to work.
Shawn: So with 17M people, you must be mostly talking about 23andMe type data – non-sequencing based. Is that right?
Dawn: Correct – 23andMe, Ancestry, MyHeritage to name a few. There are also products like Color, focused panel products, Veritas products, and Helix. So, yes, the current state of direct to consumer is largely microarray, but it’s evolving towards sequencing as well. Secondly, there’s any kind of genetic information that’s generated in the healthcare system and a lot of individual directed, physician mediated exomes and genomes starting to occur. Lastly, the third would be individuals who are participating in population scale research. The aim is to give those individuals data back and they can then come into Luna with that data. So those are kind of three ways that somebody would come in with a DNA profile.
The second (bucket), the clinical medical – you have the opportunity to grab that data from your healthcare system. And the third is more future looking – lifestyle, environment, nutrition, wearable sensors, Fitbit type stuff. So that’s the three types of buckets and how we would get information from those three categories.
Shawn: So why is the third category more for the future? Because the technology there seems to be the most mature.
Dawn: I just meant “future” in terms of the versioning of our product. You’re right, the data is out there. Just when would we be taking that in? That’s probably a couple of versions out.
Shawn: Getting 23andMe data sounds easy – I’ve downloaded the raw data for my whole family. But getting EHR data sounds hard – I wouldn’t know where to start. Will your customers know?
Dawn: There are lots of companies out there, we call them clinical data aggregators, that can do this – PicnicHealth, to Apple Health Kit, to Zipdy, which is a local company. There are many companies building elegant products to make acquiring DNA data super easy for you – by just plugging in a few healthcare systems you’ve gone to and they’ll go out and grab it too to maybe just offering the framework for you to store that information. We don’t need to reinvent that API. We’ll largely have a system that’s compatible with a lot of great products that are up and coming in that category. But it’s going to be really important. We’re mainly concerned with the format that those products pull in and serve up the right calibration, harmonization, and structure of the data on the back end. Everything we bring in has to be valuable to researchers and so that kind of data structure is really important.
Shawn: So you’re not going to generate any data yourself. What if a consumer comes to you and says, “I want to do this, but I don’t have any data”?
Dawn: I’ve been thinking about that question a lot – “I’m interested in Luna, but I don’t have a DNA file”. We are exploring certain partnerships right now. Our goal is not to be exclusive and to have lots of partnerships with data generators, but we haven’t made any commitments in that category. But we do know we don’t want to be in the business of collecting specimens, generating data and interpreting that data. There’s a robust marketplace already that we don’t need to reinvent.
Shawn: How big does this have to be? If you put this out there and you get a few thousand people signing up, is that big enough to do anything useful?
Dawn: Yeah, it’s interesting. It depends on the question you’re asking. So for example, we are engaged with foundations that have a charter to tackle disease, they have a budget to do so and they have engaged patient populations. What many don’t want to do is host a genotype phenotype database and engage people longitudinally to collect health data. So we come in solving a problem. If a foundation comes in with a very enriched patient population that has that longitudinal data, perhaps a few thousand is rich enough. It just depends on your research question. The problem we want to solve is the scope and scale of research databases has been lacking in addition to the longitudinal nature of the data that’s out there. And in that regard, the largest human health database is a great goal for us to aim to.
Shawn: What are your ideas for engaging consumers to gather that longitudinal data, to get them to constantly logging into your system?
Dawn: So we think one of the issues in individual engagement and medical research has been the lack of transparency in terms of their role and the role in the research as well as a lack of transparency in the reward stream that’s coming from this. There’s a healthy tension especially now with the notion of economies of sharing – I have data, it’s valuable; I should share in any rewards that come from that data. And so where we’re really proud of in this model is that we are going to engage people transparently and fairly. We want them to understand the type of research that’s being done in the database. We want to engage them with utility when they’re in the system and we will flow back proceeds of the database. And we’ll think of other ways to be engaging, but we think being fair, being transparent, being mission driven, will be very attractive for individuals who want to participate in research.
Surveys are a great way to capture additional data. Looking ahead, What could this look like as technology just operates in the background of us living our lives because it’s just on our wrist? You could imagine contributing a life’s data to research just after one consent, just opting in for a device or a sensor. But obviously in the meantime we’re not there yet. And so certain surveys and certain look-back mechanisms in terms of pulling EHRs (will be the short term method). But as we move forward, I think the contextual information of lifestyle, environment, and nutrition is going to be extremely exciting. And you can’t collect that in a one off basis or one research protocol. That’s why engagement matters. That’s why a people-centric community database matters because you have this living, breathing longitudinal ecosystem.
Shawn: You mentioned gathering data from foundations. Do you see Luna DNA being used by groups, like in a co-op, as well as individuals?
Dawn: It’s interesting because DNA is the ultimate unifier. We’re all so much more alike than we are different, but the opportunity to create teams and affinity groups (in the database) is super exciting when you think about discovery. And that’s why having a public benefit corporation structure, having a community owned (database), using the best technology to deliver transparency, having a great team that spent their whole entire career making genomics data and now saying, OK, so what, what are we going to do to make it more actionable? Let’s get in the business of really turning those data into actionable insights. It’s just a great time to be in this space.
Shawn: Do you envision it truly being a group effort in that way or do you imagine people being rewarded individually somehow? Like if they have more valuable information or they’ve included more information.
Dawn: The problem we’re here to solve is that the scope and scale of research data has been a bit hindered. We haven’t had the size of data sets. We haven’t had the depth of omics. We’ve lacked a lot of the contextual information like lifestyle, environment, nutrition and what we have had has lacked diversity. And so thinking about inverting that model and inviting people in to bring all those pieces of the puzzle together. Foryour question about value, there are at least three types of research that can be done. You can look at the broad metadata and ask questions about the broad data set. You can also through indexing that information do more focused studies – women over 40 with a specific genetic profile, for example. And then there’s a third set which is more directed research, which can be enabled through smart contracts. An example might be “I’m a researcher, I’m looking for this type of profile and I want to reach out to those individuals directly and have them fill out this survey or maybe explore this diet.” And so in that last model there is potential for other people to generate more value in the ecosystem than the broader co-op. But for the most part, your value in the co-op comes from how much you donated. The more you donate, the greater percentage ownership you have in that co-op.
Shawn: How will you be compensating participants? How will they share in the rewards?
Dawn: It’s a community owned framework. As value is derived, it’s going to flow back to the individual members who contributed the data. That’s a very unique aspect of this business model.
Shawn: So let’s talk about the elephant in the room. When Luna DNA put out the first press release, blockchain and the cryptocurrency “Luna Coin” were part of the business model. Is that still the case?
Dawn: The strategy remains unchanged, which is a human health database owned by the community. Value flows back to data donors when research is conducted, consistent with the charter of our public benefit corporation. So none of that has changed. Some of the tactics have changed obviously with a lot of changes in the SEC environment. We are pivoting to a much more traditional approach to ownership that we will be announcing soon. But the overall business strategy and mission remains the same.
I hope you enjoyed our look at Luna DNA. While they aren’t accepting samples just yet, you can join their newsletter to keep up with their latest announcements. Be sure to keep an eye out for future entries in this blockchain genomics series.