Your DNA, Your Say

We're being told that our DNA can uncover many secrets about ourselves from unlocking answers about our hope to reconnecting us with long lost relatives. But since DNA information is stored in online databases, it can end up traveling further than we could ever imagine .

Once the DNA has been sequenced, whether it's in the private labs of companies like 23andme or ancestry.com or in clinical settings for healthcare, medical research or law enforcement, it's almost impossible to know where in the world that data ends up, who may be looking at it and what they may be using it for the global nature of online data is exactly why the, your DNA, your safe study was developed.

It's a survey designed it to gather public perceptions on DNA, big data concerns around privacy and who people trust. Well , it was designed by researchers in the UK. It was also translated into French and Polish. So public attitudes from different countries could be compared, but interest in the project quickly spread as other researchers around the world, volunteer to get involved. As a result of the survey has now been translated into more than 15 languages conducted across five continents.

Today, I'm talking with the creator and lead researchers of your DNA, your se , and the Middleton and Richard Milam from the welcome genome campus at the Sanger Institute in Cambridge. But what the study is uncovering on global perceptions on DNA sequencing, data ownership, and people's concerns around trust and privacy in a global digital,

Hi. Well, hi. So I suppose by way of introduction, if we could start with a little bit about yourself, how you wound up in the field in the position that you're in and what kind of drives your interest in the work that you do.

So that's an , a nutshell what we do. And my part of it is leading the group.

So some of this variation will give us eye color and our hair color and our disease profile and our , you know, everything really about us has some biological component to it. And that is controlled by genes and DNA. Genomics is the study of that, but genomics is very much a big data science because we all have 3 billion bits of DNA. So the scale of it is enormous. And so no one single scientist or research Institute or project is actually going to unravel the meaning of , of genomics .

Um, it has to happen via collaboration because the volume and scale of the, of the issue at hand in the science is so enormous that it needs to be done through collaboration and collaboration involves sharing of data.

And within that, there was always a commitment that was driven by people like John Sulston at Sangha to make the data that was generated within the public initiatives available for the benefit of humankind as widely as possible, and as soon as it was generated . So there's been this debate since the early nineties around making genomic data available for wider benefit.

It may be connected to their personally identifiable information such as their name and their date of birth and their address and things, or it may be disconnected from that. And so we're really fascinated in terms of the level of awareness about this, and whether people feel they want to have some say in how the data is managed and processed and shared, and the privacy issues around it and what they feel about some of the ethical and legal issues that it raises.

And you might say, well, why is this even relevant to anybody? You know , and that's a really good question. I mean, the reason it's relevant is because even if you, as an individual have never personally had any level of genetic testing yourself, you know, blood donors , um , take part in DNA research. So if you take the bottom of the consent form and said, I'm happy for my DNA to be used in research.

So even if you've ever done that, or if you've never had a genetic test as part of clinical health care services, or if you've never taken part in an ancestry test, or you've never done a 23 in the online direct to consumer test , if you've never done any of those things may think, well, this isn't relevant to me, but the thing is, you're going to be biologically related to somebody who has even distantly even amongst your family, that you may not even know.

So the issues of relevance have moved beyond the individual, into the family and now into wider society. So it really is time for people to kind of wake up to the fact that this is happening. And so our genomic data, even if it's not how it was personally, if it's our relative stated , if we think about all the ancestry testing that's being done around the world, that data that can connect to us is out there. And we all have so much online data about us already out there.

So whether we're on Facebook or whether we do Amazon shopping or whether we do online banking or whether we just email people, you know, we may feel that we don't have a massive imprint, but actually we do. And there are algorithms that can connect all of that and say , if you bring genomic and health data into that picture, what does that do? What do people think about that? And that's, that's what we're researching is trying to get to the heart of that.

So it's within things like healthcare initiatives, like the UK , a hundred thousand genomes project related to those it's generated within research projects, but we're also increasingly producing genomic data within commercial and corporate settings and using it within those settings as part of drug discovery and drug development.

And so there's a , a persistent question in terms of kind of social and economic aspects of genomic data in terms of where do we collect this data who collects it and who's using it? What is it, that's the benefit derived from it? And how do we, how do we ensure that it's as widely shared as possible

And if we are in partnership, what does that actually look like? And, you know, is it okay and acceptable that you can buy this data and sell this data and make profits from this data, the people whose data it was to start with, what do they think about that? And are they actually fully aware of the journey that that data takes and these sorts of conversations about, are you okay with that? And if you are, then what does good look like for you?

And if you're not okay, then again, what does good look like for you? What do you, what do you think should be done with this? How should we as a society be forming the governance structures that help people feel that they're protected and that, that privacy is protected. And yet at the same time, can we advance human health and understanding and help to treat and cure diseases?

And how do these partnerships actually work and how do the partnerships between non-profit and for-profit industries work? And what's the public voice in all of that .

I can see how that is going to influence a lot of perceptions about the ownership of data and the risk that it poses to health . There's a situation where 99.98% of genomic data is identical between individuals.

So there is a relatively unified human genome, which I think, I mean with important distinctions that I think we might come back to later, but the way in which that genome translates into care into the genomics industry, into a wide range of products and services, isn't necessarily related to that human uniqueness, it's related to the diversity of health systems and social political and economic contexts in which research and healthcare are funded and delivered.

It's a really important reason for understanding how members of the public patients, people who are donating their data and people who are accessing genomic health care in the future, what they think about the whole system and how their values and interests are reflected and represented in the systems that are developed around genomic.

So the data sets that are guiding how to interpret Genomix globally are just heavily biased, lack of diversity in data, but also a massive disconnect between the way genomics has evolved and is delivered now in different countries. So America, particularly with the, you know, the, the commercial kind of interest and the different healthcare settings to many of the places around the world have a very different cultural perspective on genomics.

And a lot of the research has been done about public attitudes comes from American researchers and American audiences. And we were very mindful that, you know, that doesn't represent the world. You know , even though they're often the ones that are leading on policy creation for the global audience, and they're very influential in the way that they write and feel that the direction of travel should be going, that doesn't actually meet the needs of, of the rest of us.

And so one of the reasons why we wanted to do the, your DNA, your say study was to try and capture some of the views of people, not only in America, but also outside of America. So what we've done is we've reached out to many countries around the world and we , in our final data set, we've got completed surveys from public audiences. So representative public audiences in different countries, giving their thoughts about genomics and data sharing. And we explain what all the fat is in the survey.

So people can come to this without knowing anything about it, but we've particularly recruited in whole diverse mix of country. So , um , India, Pakistan, Brazil, or Costa, Japan, China, all the different European countries, who've recruited in Polish and in German and in French, and also been to Australia and Canada and the UK. And we will soon be recruiting in Ghana. And we've really made an effort to translate our work into many different languages.

So at the moment we have 15 languages and see , can you imagine the translation of genomics , which is tricky enough in a native language to translate that into lay language, let alone to translate it into countries where there's no literal translation and the science isn't there yet.

So in Ghana, for example, translating genomics into tree and UA, which are two national languages, the literal translation is something like gift within the family is, is what DNA genetics means gift within the family. I'm pretty sure that's also the slogan for ancestry.com. Yes . It does sound a bit like that, doesn't it? Yeah. Yeah. So it's, there's loads of sort of different way , ways of framing, what is DNA and what it means to people.

And this is what we want to do understand, you know, his DNA, something that means disease and health. Is it something that just means family and inheritance? Is it something that means police and crime and identifying bodies, which is actually what it does mean for a lot of people, if they've heard of DNA profiling in databases to do with crime and things, is it to do with climate change and biodiversity, genetically modified foods.

We actually genuinely don't know because that kind of global research actually asking people hasn't been done before. So that's, what's so exciting about this particular project that we've been running is that it's the first to really try and gather on a massive scale in lots of different languages to try and get a sense of what, what people understand by all of this and what they're familiar with and what they trust.

And what we found is that broadly across the world, outside of America and Italy, interestingly, people have really not heard of genomics. And they're very unfamiliar with what's actually going on in terms of genomic research. You know, they're not aware of this partnership between non-profit and for-profit industries. They're not aware of energetic testing that's happening and the relevance of it to them. They're just very unsure and , and , and knowing about it all.

And then of course, if you don't know, or understand, or you're unaware of something, then you're more likely to be suspicious of it or mistrusting. And we found that that link as well. So people are more likely to say, well, no, I wouldn't trust lots of people with my data. I , the data sharing process, and she's absolutely fundamental to do they mean research. And I know I wouldn't be willing to donate my data for this kind of thing, because I just don't understand it.

And I don't, I just don't know enough about it to feel trusting of the process. So that's what we've found fairly consistently around the world. I mean, there are obviously different sort of nuances between certain countries, but broadly speaking, that's, that's the direction of travel and actually incredibly valuable to know it . It just shows that we need to be helping people to come up to speed with this because it's happening and it's happening on a massive scale.

And there's a lot of commercial interest in this particularly say in the African genome and certain particular communities, the genome of certain diverse groups, because I said before, you know, the data sets that we already have are incredibly biased to white those in European ancestry. So there's this big push at the moment to involve all sorts of different diverse communities in genetic research. But hang on a second, are they ready for this?

Are they up to speed with what it means and what their contribution will offer? And do they want anything return? I mean, that's not the big question is what does reciprocity look like

So therefore we need to start thinking about screening you for breath , for example. But if you found exactly the same gene fold in a patient from Zimbabwe, you could be giving the wrong information. You know , you could be saying you've got a high risk of breast and ovarian cancer, but that's because the information that guides that data is very, very biased.

And the research has not been done on that in Barbara in population, does it may well be that the variation they have is actually completely benign. So there are quite serious health problems that, you know, you either pick up or you miss, or you misdiagnose, or you do the wrong thing with. So there is a bit cool from trying to do good from a health perspective, to get more diversity in the samples and the data, and just get a more solid, robust way of interpreting the data.

But yes, of course, there's going to be actors that don't have that as their motivation. And, you know , we're always incredibly shocked by the number of online companies that are popping up all the time, claiming to predict all sorts of things, which are just a complete nonsense from your DNA, like a certain wine, for example, with it , your child is going to be better at football or not, or more worryingly the profiling of children for certain educational settings.

The science behind it is poor. The societal impact of it is not fully understood. There's quite a, in my view risk of doing quite a lot of home,

So actually things like ancestry.com 23 and me or any of those that have been far more early to market, I suppose, in the United States and far more heavily marketed in the United States, there is a question of how much of that familiarity is being driven by, by that .

And so, you know, there's this massive web of connections with people that really are so distantly related, you know, certainly from a genetic standpoint, my own experience working in the NHS and we wouldn't call them related because they're so distantly related, I can give you an example. My , my dad , uh , was contacted by somebody from New Zealand who had got his details through an ancestry websites . And , uh , she was like a 16th cousin , 10 times removed or something.

So the amount of DNA that they shared was so small, but because it was framed as I'm , I'm a long lost, distant relative. My dad sent her via email, our entire family tree, and he says to me, but it's family. And I said, dad , I just thought , my God, all that really personal information is this now left? Maybe it's fine. Maybe it will be fun. And I think that's the thing that we think, well, what , what can happen?

And that's the thing that we're interested in, but what can happen is that too very easy to connect that, to have a few personal pieces of data, you just put that in a Google search and you found that person online, and then you can very easily, if you've got the know how connect that to their genomic and health data as well. And most of that, of course it should be completely confidential course it should, but not always and not all the time.

And so we do say, you know, now in your new research studies, you know, part of the consent process is explaining, you know, the absolute intent of this project is to keep your data fully confidential, anonymous, whatever, but we've all got so much data online about ourselves. There is a real risk it's small, but you could identified from this data and you need to weigh that up. You know, what risks are you prepared to accept?

And it may well be that you feel okay taking part in a study on breast cancer. Fantastic. I'm prepared to accept the level of risk that I could be identified in the process, but taking part in a research study on ancestry that then links to health data. I would be less in favor of me personally, taking part in that because I think I could be identified and then, and then what could happen. And so the research that we've been doing is exploring well, what do people think could happen?

What are they worried about? And so we've been finding all sorts of interesting things, the theme of governments, knowing things about you, that you haven't chosen to share has come up , um, thoughts about, well, could my DNA be copied and planted at the scene of a crime? I mean, that sounds very, very scifi, but could it happen? What's the sense of governance and control and protection. And am I safe against that? Is there a risk of that?

You know, what about sharing information and families that, or relatives , names , things that I haven't chosen to tell them? What about market research companies now targeting me for certain campaigns because they think I've got a predisposition to diabetes, or what about insurers? You know, they now go to discriminate on the basis of , of genetics. And of course they shouldn't. And in many countries there are legal protections in place, but they're not completely fail safe.

I know for example, in the UK, it would be against the law to discriminate on the basis solely of a genetic test result, but you can discriminate on the basis of a family tree and the family history. If you've got cancer in the family, you have to declare that and policies and weighted according to that.

So there's sort of lots of gray areas around this, and we're just not there yet in terms of, well, certainly not global governance of this, but even country by country, the governance is sketchy in certain places, and that's why we need this kind of global conversation to feel that we can be part of deciding these sorts of things because it's happening so fast. It's very, very now it's, it's it's happening, you know, it's time for everyone to catch up.

Cause we're just sleepwalking into this minefield of data. And the worry is that, you know, nefarious actors will just decide what it's going to be. It's going to be Google and Facebook that make decisions about it. And, you know, we need to be thinking wider than that.

It's often one of the best ways of predicting whether somebody is at increased risk of cancer is to look firstly at their lifestyle. But secondly, more importantly, their family history. So if they've got lots of relatives, who've died young in their thirties , say of breast and ovarian cancer, then you can be fairly sure that there is a single gene that's responsible for increasing the risk of breast and ovarian cancer.

Whereas if you find that same gene fold in another family that has no history of cancer, then you're less certain of what is it going to play out this cancer or is actually, is this something modifying the behavior of that gene? So actually one of the best ways of determining, you know, disease is actually to look into the family.

We do know that if you have say a very strong family history of young onset cancer, and you have a genetic test that shows you've not inherited the family, gene that's causing the cancer in the family, you are fully entitled to let the insurance company know that. So you've not got with them in your predisposition and if that's been confirmed and they should take away the history waiting that they would have previously applied. And that's what they say they do.

Of course let's hope that is what they do.

That could also be the output I suppose, of putting it to corporate partners or private partners of here is how you can increase public trust or transparency in your work. So people are more likely to use your services and more likely to trust you.

So we've been, we've been looking at that and , and particularly drawing out the kind of emphasis of being clear about when people are collecting and using and sharing data, who is it that's going to benefit from that and how, how are those going to benefit whether not just in terms of companies benefiting, but also, you know, clinicians and scientists in terms of the personal and institutional benefits of, of being able to do that research and , and being kind of transparent about those

benefits and then things like as well, like protecting things like the ability to withdraw your data from research, which can be very complicated. And particularly once data started to be used. And if it's been shared, it becomes very difficult to track where that data has been used.

So then it's a case for those bodies that are responsible for using data companies, researchers, whoever of being clear about the fact that, you know, once it's been shared, we can't necessarily protect that rights , withdraw and being upfront about that. But consequently being clear and transparent again about who it is that's using that data and why and what they're going to do with that.

So are some of the things that we're trying now to feed back through various different channels, including head of international genomics initiatives and things like that.

They are an umbrella organization for 600 smaller organizations. I , I say smaller that they're all very afraid pig. Um , so the second Institute welcome trust , um , the Brode Harvard's Melbourne genomics Alliance, the department , do they make sense that, you know, all the different Genomix players across the world as part of the global Alliance, which you don't mix in health.

And so we are feeding the results from our research into policy initiatives to coming out from that, but that isn't enough. You know, we need to be taking every opportunity to be talking about what we're doing and the outcomes of the study, and also engaging directly with media and promotion off the work, because we want to actually shift something at the very, very basic level. We know broadly that public audiences don't know what any of this is.

You know , they're not having conversations in the family about what it means. And so we need to collectively figure out how to help that. And it's not about , um, you know, education and knowledge and , you know, sort of really pushing genetics down people's necks. It's not about that. It's about thinking creatively about how you shift awareness in a society, in other projects that we're running, we've been doing , um , work with the creative industry.

So storytellers, filmmakers, animators, advertising people, marketing people, really China to start, how can you make this click for people? What metaphors can you use? What means can you use, how do you make this relatable so that people understand why it's of relevance to them and why it's going to be part of their future lives? So, as I say, even if they've never personally engaged with it yet, it will become much more of a part of particularly part of healthcare .

So , um, you know, the ethical issues and this societal issues are on, are relevant to all of us. And particularly in the UK, for example, the new genomic medicine services launched in 2018. So this now means that pretty much any health professional across the NHS can now order genetic test is moved out to specialists critical services, clinical genetic services, and across the whole of mainstream medicine.

So these kinds of issues of what does this mean for me and my family and my relatives is now relevant for anybody engaging with the NHS there, whether that's in obstetrics or fetal medicine or in cancer services, or in dermatology, Oriente or radiography, you know, genetics will be part of that. And that's not just related to you in your own interaction, it's related to everybody that you were linked to as well.

So we, you know, we do feel this pressure of getting the messages of our research out very widely, and we need to think very creatively about how to do that and also do it in an evidence-based way that works. Um, so then that brings us back to the research question. How do you do that?

And actually I think the, the onus and actually probably the place where actually a difference is more feasibly made is in terms of turning it on to the risk onto researchers and companies working in this area and thinking not necessarily about, yeah, how can, how can we make people trust us, but what can we do in order to be seen to , to merit people's trust?

Um, so I think that this does come back, and this is something that you see in wider debates around the use of data as well at the moment, as well as in the context of genomics, it is fundamentally things about being transparent in terms of what data you're using, who does that come from? What are your plans with it and how do you look after it? And then also being clear about when you use data, who's going to be benefiting from that.

And if it's being used primarily for if a private benefit, then, I mean, that needs to be clear as well. And people need to know that before they enter into an arrangement where whereby their data is used. Yeah. Those questions for, for researchers, for institutions , um, for funders as well for science funders and for companies are really key.

And it's too glib to say things like, Oh, well, you know, you'll be contributing to improving health of humankind or, you know, helping to get a cure for cancer. People want to know specifically. Yeah. But is this going to help your career? Are you going to be writing more papers? Are you going to get more funding? Are you good to be personally advanced in some way by this participation? And then, then I can decide if I'm going to either give you my data.

And I think that's the real message that we're getting from the research is that people humility of our scientists and partnership and to be seen as, as yes, as partners together in solving these really big kind of world issues. I mean, historically there's been so much focused on taking data from people and leaving the scientists , you know, just to go do good stuff with it and just trust me with it all, you know, you won't understand , um , I'll just get on and do great science.

And I think that that's now very outdated in terms of thinking people have to earn trust to be worthy of it and maintain trust. And that involves a lot of a shift in terms of energy and focus as not , not enough of that at the moment within the science

industry.

I mean around, so if you look at initiatives like the kind of flagship genomics initiatives in the UK and the USA, so the a hundred thousand genomes project and the all of us initiative in the States have made a huge effort to actively involve the communities whose data are being both genomic and clinical data are being used within those studies. So that they're really trying to involve people in a model that , that gives them a more active role than just a kind of broadcast.

This is how we're using your data.

That's a really encouraging trend as genomics, I think tries both in the kind of private sector, I think, where you start to have partnerships between companies like 23 and me and the GlaxoSmithKline, it becomes clear that, you know , the value of the data is not always in what you're getting back as a consumer, but actually what it is that a company can extract from that, that to some extent, when you enter a relationship with 23 and me, and you send your sample off, you're aware to a degree

that this is a commercial transaction. The challenge then is when you for public sector bodies, when they enter into relationships with pharmaceutical companies or diagnostics companies that are doing medical innovation.

And that , that are kind of key part of how we currently as a society, do medical innovation, that when a public sector organization entered into a partnership that it's, that that's always really clear and that this is something that's that people are informed about and people have a chance to talk about upfront rather than it being seen as something that's, you know , slipping in by the door .

And you know, that the company doesn't actually make the money from you buying the test, if they actually make the money from them selling your data wrong . And so, you know, it's going into these kinds of relationships with companies, with your eyes wide open and understanding that that's actually the model that you're engaging.

Um, and there's often a mixing between recreational genetics as it's called, you know, isn't this fun to find out if I'm a sporty type, I'm a sprinter, or I've got stamina and mixing that in with medical genomics, you know, if I got a predisposition to heart disease or diabetes, whatever, or the two often get kind of merged together, and it's quite a minefield when you're engaged in one of these tests where you're , you're going into it for a bit of fun and because it's a Christmas present and

you can think of anything else to buy your mother. And then she discovers that she's got a predisposition to outsiders, you know, suddenly it's Whoa, okay, I'd have no preparation that that was coming. How do I talk to my family about it? Is this a real predisposition to the other thing? It looks very meaningful when it's in black and white on a computer screen, but actually what is the science behind this?

Is this a real risk, you know , is this a clinical risk that I need to actually worry about or not? And it can raise all sorts of massive concerns for , for the individuals. So there's so much there in terms of the commercial direct to consumer testing companies that really needs to be tightened up in terms of protecting people.

But also when we, when we've asked people about the relationship between things like , uh , kind of legal protections for data and trust, when we did an analysis of the responses to our study from the U S Canada, Australia and the UK, the people who were most likely to value legal protections were also those who are most trusting.

And it's a difficult relationship to unpack , you know, where they trusting because of the legal protections or converse the, are those people who are already have faith in the system that they're giving their data into. Are those people who are more trusting, also more likely to then trust a law that, that same data system says we'll protect their data. So there's a chicken and egg question there.

And actually , and when we then asked the question about what kinds of things would people see as characteristics of trustworthy data systems that the importance of knowing about the legal sanctions for misusing data actually came quite a way down the list. So in some countries that was more important than another . So in Japan, for example, it came out very high, but generally it was less important than kind of questions about transparency. And again, there's a chicken and egg thing.

Is it something that's down the less, because people are aware that it's there and so they don't necessarily need to be constantly reminded about it, or is it just that the, these other kind of features of the interactions that you have with the people that you're giving your data to use? Is it the fact that those interactions are more important than the legal framework?

So I think, yeah, even in Europe, that question of the GDPR is , um , what its impact is on, on public trust, I think is, is, is a question that's really fascinating to explore a bit further. And then just to come back to Anna's point about the kind of recreational genetics and the kind of things that you might find out over, over Christmas that you didn't necessarily expect to find out. And one of the kind of classic examples is kind of finding out that your father isn't your father.

And that kind of goes, I think back to the, to the question where we started about why is this really important information to be thinking about? And that , you know, when the direct consumer genetic testing companies started up, a lot of them had taglines like know yourself and this idea that there is something about genetics, that's a kind of core part of your , your identity.

And that then when you kind of go into the questions of paternity and genetically inherited conditions as well, that it's part of your family identity. When you talk about the things that we mentioned earlier about ancestry testing, that it becomes these questions of the groups that you belong to.

So you know who you are as an individual, how you sit in a, in a family or a kind of kinship group, and then how you set as part of a broader social group, all these things that , that get wrapped up into, into how we think about genomes and how we think about using the data related to them and who else is involved in those discussions.

Because to think about it as solely an individual choice and an individual's decision when it relates to so many other people becomes an increasingly difficult position to maintain

What is my place in a very challenging society to live in, have a piece of paper say, here is where you're from. Here are your people. Here is an identity. Then you have , there's a lot to latch on to .

the best way to reach us and interact with us and to follow what we're doing is to just do a quick internet search of society and ethics research in Cambridge that will pull up our website. And then the website is really an enormous catalog. All the projects that we've been working on and all the different outputs and impact of the work.

Because as I say, it's , it's one thing, writing lots of papers and doing presentations at conferences, but it's another thing trying to actually genuinely shift something and have policy impact and make changes to societal kind of attitudes and values. And so we're working quite hard on that. So the website is probably a good place to start just to see some of our work, if you're interested in and always reach out to us.

Cause we're always very, very excited to connect and collaborate and be in partnership with society on , on genetics.