Why public genomics is not a purely personal decision

I knew I wanted to be a part of Genomes Unzipped from the very first day Daniel told me about the project. But the decision to actually participate was more complicated.

I’ve spent a lot of time over the past five years thinking about the benefits and risks of personal and public genomics. By sheer good luck I became an advisor to the Personal Genome Project (PGP) in its early days, long before PGP-1 (George Church) was joined by so many others (the arrival of the PGP-1K was announced yesterday) interested in exploring public genomics. As a result of that connection, I have been able to continue to work with the PGP, exploring the issues raised by public genomics research as they arose, often for the first time, in connection with that project. And as part of my day job I edit an online publication (the Genomics Law Report) and advise clients—both focused on issues arising at the intersection of personalized medicine, genomics and the law.

As a result of all of this time spent thinking and writing about personal and public genomics, I’ve come to know a few things. I know that I believe the benefits of public genomics, for both science and society, to outweigh the risks. I also know that I might be wrong about that, but I know that I’m comfortable accepting that risk, too. I know that I want to support personal and public genomics projects. Most of all, I know the decision to join a project like Genomes Unzipped is not mine alone.

Before agreeing to participate in Genomes Unzipped, I knew I would need to seek—and receive—permission from my family. There was no legal requirement that I obtain their consent before joining the Genomes Unzipped team and, ultimately, publishing my genetic data for the world to review. But that did nothing to change the strong personal obligation I felt to seek my family’s consent.

While my genetic information is personal to me, it could also have something important to say about my family members. The strong likelihood was that it would not, particularly once you diluted the limited predictive value of most genetic markers with a hefty dose of uncertain utility and a large helping of additional uncertainty thanks to the fact that I share, on average, only 50% of my genome with each of my family members. But while I could calculate that risk as small, and accept it for myself, I could not unilaterally accept it on behalf of each member of my family.

Knowing that, I set out to talk to my family about Genomes Unzipped and my interest in participating in the project, and to learn whether that was a decision they would each be willing to support. Spoiler alert: there is no surprise ending here. You would not be reading this post if any member of my family had not fully supported my decision to join Genomes Unzipped.

Still, I learned a great deal from the process of seeking informed consent from my family for my participation in Genomes Unzipped. And as the cost of genomics sequencing continues to decline, and more genomic data inserts itself in more ways into more of our lives, conversations like this will become more common. More of us will want, or need, to talk with our families about genetics, including the potential risks and benefits of sharing our genetic data with others (and, for some, with the whole world).

Last November, I sent the email below—which has been reproduced verbatim and with permission, with only names and email addresses redacted—to my parents , brother and sister shortly before we gathered in North Carolina for the Thanksgiving holiday. I hope that it, along with this post and several more that will follow in the coming weeks from me and my Genomes Unzipped colleagues, will serve as a helpful starting point for a discussion about how to talk and think about familial informed consent in the context of public and personal genomics.


From: Vorhaus, Daniel

Sent: Monday, November 23, 2009 2:00 PM

To: [Mom]; [Dad]; [Brother]; [Sister]

Subject: A family discussion about public genomics

Dear Family –

We’ve talked in the past about my work with the Personal Genome Project (PGP) and the possibility that I might be asked to participate in that project at some point by making my genetic information publicly available for research purposes.

I still have no plans to publish my genetic data through the PGP.  However, I have been advising a different group of researchers (in Cambridge, UK) and they have asked if I would be willing to publish portions of my genetic information through that project (which is called “Genomes Unzipped”).  I’m writing because I would like to say “yes,” but I first want to make sure that each of you is OK with the decision.

Below I’ve attempted to describe what I would be doing, why I’d like to do it and what that might mean for each of you.  It’s a little bit long, I know, and I apologize for that.  If it’s any consolation, knowing me it could have been much, much longer.  I’m hoping that you’ll each be able to find time to read it and that we can talk about any questions or concerns that you have when we’re all together this week.  In any event, I will wait to speak with all of you before I respond to the team from Genomes Unzipped.

Without further ado…

I  What genetic information would be published?

The information that I would publish, at least at first, would be information generated by the company 23andMe (http://www.23andMe.com/).  You can see a list of the traits for which they generate information here.   It’s important to keep in mind that the information generated by 23andMe is not a complete genome sequence (all 6 billion base pairs).  What they do is genotype (or “read”) several hundred thousand letters in an individual’s genome that are thought to be associated with certain genetic traits.  Some of these are associated with diseases, like cancer, while others are for much more benign conditions, such as whether you have detached earlobes or why [Brother] thinks cilantro tastes awful.

At the moment, 23andMe is probably the most-recognized consumer genetic testing company.  They claim to have some 30,000 customers, and some of them have already made their information publicly available.  (If you are interested in seeing examples of this you could go meet Geoffrey and his genes or go look at this graphically-unfriendly list of public genomes.)

At the moment all I’d be receiving and publishing is data from 23andMe.  But I think it’s probable that in the next few years I’ll have the opportunity to publish my entire genome so, for our purposes, that’s really what I’ll focus on from here on out.

II  Why does this affect you? (Part I: Genetics)

We’re related? Are you sure? Because of your relationship to me, I share approximately 50% of my genes with each of you (but I want them back!).  To put it simply, what this means is that any genetic information revealed about me will say something about you.  But what?

In almost every case, what my genetic information says about me will be a probability (for example: genetic variant X is thought to contribute to an 18% increase in lifetime risk of developing diabetes).  And, in almost every case, what it says about you will be even more probabilistic (because you are my parent/sibling, you have a 50% chance of having a genetic variant X that is thought to contribute to an 18% increase in lifetime risk of developing diabetes).

Most genetic information is probabilistic for several reasons, including the fact that many genetic factors remain unidentified or poorly understood.  Even more important is the fact that most traits, whether diseases or other characteristics, such as height or high IQ, are influenced by a complex combination of genetics and environment.  That means that, most of the time, no matter how much you know about an individual’s genes you won’t be able to predict what traits they will develop with anything resembling certainty.

Huntington’s (and other scary genes). Of course, there are exceptions to everything I just said.  For instance, it is possible that I could learn something in the future that would also tell me something about one or more of you with much more certainty.

Here’s probably the most extreme example (although it’s also extremely unlikely): I find out that I am homozygous for Huntington’s Disease, which is a dominant monogenic degenerative condition.  How about we try that in plain English?  Huntington’s Disease is a disease where, as you age, the nerve cells in your brain deteriorate.  It’s a truly terrible disease to suffer from and it’s caused by a mutation in a single gene (that’s what “monogenic” means).  If you have even one copy of that gene you have a 100% chance of developing the disease if you live long enough (that’s what “dominant” means).

So, if I had my genome sequenced and found out that I had one copy of the Huntington’s gene, what would that mean?  First, it would mean that I would develop Huntington’s at some point (symptoms usually start in middle age) unless I either died of other causes or unless a cure was developed.  It would also mean that each of you had a 50% chance of possessing the Huntington’s gene as well.

Now let’s say that I somehow wound up with two copies of the Huntington’s gene, which can happen in rare cases.  That means that I received one copy from mom and another from dad, which means that both mom and dad would be Huntington’s carriers.  At your age you should expect to see symptoms soon, if you weren’t seeing them already.  (Of course, the fact that neither of you have the symptoms of Huntington’s is a very, very good indicator that you’re not carriers, but this is a hypothetical.)  It also means that, [Sister] and [Brother], your chance of developing Huntington’s would be at least 75%.  That’s something that you might not want to know if there’s nothing you can do about it.  Then again, you might want to know so you can be on the lookout for symptoms and plan for the disease in advance.

Why spend so much time talking about Huntington’s Disease?  Because, in many respects, Huntington’s is a worst case scenario.  It is determined entirely by your genes, it is fatal and, for now, it is incurable.  And when thinking about risks it’s important to try to understand the worst case scenarios.

Then again, it’s also important to remember that Huntington’s is not typical.  Even other scary diseases that we think about (including Alzheimer’s, heart disease, cancers, etc.) lack at least one of the features that makes Huntington’s so frightening.  Alzheimer’s, for instance, is not deterministic in the same way that Huntington’s is.  Even if you have the worst combination of Alzheimer’s genes, there is still a reasonable chance that you will never develop the disease at all.

Ultimately, it’s important to keep in mind that the vast majority of what I’ll learn about myself by sequencing my genome is likely to be benign and/or uncertain.  And because we each share only 50% of that genetic information, anything that you learn about yourself is almost certain to be even less certain.  But there are exceptions, and I hope that I’ve given you a sense for those as well.

III  Why does this affect you? (Part II: Discrimination and Other Risks)

Of course, even if I publish my entire genome for the world to see, that doesn’t mean that you need to be one of the people looking.  In some ways this is no different than what would happen if, for example, mom went and had a genetic test for certain breast cancer genes because our family has a history of breast cancer and her doctor decided it was a good idea for mom to be tested.  She’d find out her results and then we’d all have to decide how to share and discuss that information within the family.

The big difference here is that, by publishing my genetic information online, it’s possible for other people to access that information and to use it in ways that you might not like.  That could include telling you about what my information says (or what they think it says), and what that implies for you and your own genetic information.

There are other potential risks to consider as well.  This could include your being subjected to negative or upsetting comments or attention from friends, colleagues or even complete strangers (“Dan has bad gene X. He sucks!” or “your brother/son has bad gene X. You suck!”) or simply having to spend your time answering questions about my or your genes (“How do you feel about the fact that Dan has bad gene X?”).  It’s also possible that somebody could trace my ancestry – and thus yours as well – back to some infamous historical figures (like Genghis Khan, although that might not be much of a surprise).

Genetic Discrimination. One particular risk that people tend to worry over is genetic discrimination: the possibility that somebody could use genetic information to deny you some type of benefit or opportunity.  A few words about this: it is now illegal for employers or healthcare insurers to discriminate against you, me or anybody on the basis of our genetic information.  That’s a recent legal development and a very good thing. However, it’s not perfect protection.  Genetic discrimination will always pose at least some risk for several reasons.  First, just because something is illegal there’s no guarantee that somebody won’t break the law and, what’s more, there’s no guarantee that they’ll be caught or stopped if they do.

Also, the new federal law that prohibits genetic discrimination (it’s called the Genetic Information Nondiscrimination Act, or GINA) doesn’t cover every kind of insurance.  For instance, disability insurance and long-term care insurance (which is something you might want if, for instance, you found out you carried – or might carry – the gene for Huntington’s) are not covered by GINA, although discrimination on the basis of genetic information may still be illegal under other federal or state laws.  As a practical matter, however, the risk of discrimination to any of you is much lower than the risk is to me.  And from all estimates (as well is in my opinion), the risk to me is very small to begin with.

Unknown Risks. It’s difficult to say with any confidence how likely it is that any of these risks will materialize.  What’s more, it’s my opinion that the most important category of risk in all of this is what we can think of as “unknown risks.”  Here is how the PGP’s informed consent form (which I helped write) puts this problem:

Your DNA sequence data, trait data and other information related to you that is made publicly available by the PGP, while directly associated only with you, may also have relevance to your family members. Although in many instances any conclusions that may be inferred from your publicly available information may be speculative with respect to you, and even less predictive with respect to your family members, the complete set and magnitude of the risks that the public availability of this information poses to you and your relatives is not known at this time. You are strongly encouraged to discuss this study and its potential risks with your immediate family members. (emphasis added)

The act of publishing identifiable genetic information for public consumption is still a new exercise.  Although I certainly wouldn’t be the first to undertake this, there also aren’t many public genomes or even partial genomes out there at this point.  (The number is probably fewer than 100 worldwide, although it’s hard to say for sure.)  In the end, this is all so new that the long-term consequences of making genetic information publicly available are not well understood at this point.

As a result, there’s simply no way for any of us to list or evaluate all of the risks associated with this decision.  Just as with every decision in life, we need to do the best that we can with the available information.

(One final aside is that the risks associated with this decision aren’t necessarily limited to the four of you.  One important issue that I didn’t mention above is the possibility that I will be placing my future children at risk. Although I don’t have any yet, I still need to consider the fact that this decision could have consequences for my children – if they are biological they’ll share the same 50% of my genes that you do – who will not have an opportunity to provide any of their own input.)

Why am I doing this?

So why am I even considering doing something that poses risks – both known and unknown – to me and the members of my family, both present and future?

At a very pragmatic level, having now spent years thinking and writing about the risks and rewards associated with public genomics research, I have come to conclude that the rewards simply outweigh the risks.  I actually think this is true by a wide margin.  But that’s just my own educated guess and, even if that is true in the abstract, it’s always possible that one of us could wind up as an exception, harmed by this decision in any one of the ways described above.

More fundamentally, I am willing to assume the risks of participation – and I am asking you to assume a share of those risks as well – because I believe in the importance of scientific research to societal progress.  And I believe that neither can succeed without some degree of sacrifice and risk-taking by people like me.  I am not a scientist, but I am fortunate to have the opportunity to participate in important scientific and societal research that, although it may not benefit any of us directly, has the long-term potential to improve the lives of many.

There are countless people and professions – from the Marine Corps to the Peace Corps, and so many individuals and occupations in between – that undertake great personal risk in an attempt to improve the world around them.  It is abundantly clear to me that the risks of participating in Genomes Unzipped, the Personal Genome Project or other similar projects are trivial by comparison.

But because the risks extend to you as well, I believe that the decision to move forward is one that I should not make alone.  Before I wrap this up I want to share with you a few words from Jason Bobe, one of the driving forces behind the PGP and a close collaborator and friend, which I find particularly inspiring.

These individuals are the astronauts of our era. By sharing their genomes and phenomes and making them broadly available through participation in public genomics research studies like the Personal Genome Project (PGP), these pioneers will radically accelerate our ability to explore new frontiers of human knowledge. In doing so, they face potential risks. They are putting it on the line for our benefit and for the benefit of future generations. They deserve our support. To the moon!

All of you have supported me in so many ways throughout my life, and I would not have this opportunity were it not for that.  I hope that I will have your support in this decision as well.

See you in a few days!

  • Digg
  • StumbleUpon
  • del.icio.us
  • Facebook
  • Twitter
  • Google Bookmarks
  • FriendFeed
  • Reddit

7 Responses to “Why public genomics is not a purely personal decision”

  • Beyond the family disclosure issue, here’s an ethical problem that I would like to see discussed on this site:

    Dienekes’ blog ran an analysis on your genome data that indicates varying levels of Ashkenazi Jewish ancestry. Because you’ve made your surnames and photos publicly available, we can partially guess at your ancestry. Several of you have also disclosed that your either have or don’t have Ashkenazi ancestry. In some cases, there is a mismatch between the ancestry associated with your names and Dienekes’ test.

    Clearly, the test contains some degree of error. Most of your are scientists or have access to people who can give you clear scientific explanations, so you have some access to error interpretation for unexpected results.

    This is not so for the general public who generally have trouble processing statistical information.

    I forsee a problem where people will fully disclose their genomic data on this site, but they and the public will be limited in at least two ways in their ability to interpret the data:

    1. In some case, it will require a genetic statitician to expose deliberate attempts to manipulate your genomic data.

    2. Even where best and most ethical efforts are made to interpret your genomes, the public and some participants will not be able to understand the error inherent in the results.

    The situation is akin to the work of this year’s Nobel winner in economics: in markets (in this case, the market for genomic data) buyers and sellers do not always make contact freely. In this case, genomesunzipped is the (free) seller. The buyer is effectively the public and the scientific community.

    Because of scientific and public bias, as well as public misunderstanding, the your freely available genomic data is likely to be both misused and misinterpreted.


  • While I assume you have gathered permission from your immediate family, at which point do you and your colleagues make the cut off of where relatives are sufficiently different to no longer be considered as being likely to be directly impacted by your decision? Can you guarantee that 10, 20 or 30 years down the line this will not effect their lives?

    In this regard, how do you justify imposing your choice on your future children and even grandchildren? For your colleagues who may have offspring, do they consider it their right to give consent on their children’s behalf? Or do they have a moral obligation to act as custodians and not subject them to unnecessary risk?

    Your make the comment that there is no legal requirement for your relatives to give consent. Does this expose a danger where individuals will be forced into a situation where their genetic information is revealed against their will? Should you be establishing a legal ethical framework to prevent such a situation from occurring?

  • You didn’t mention to your family the most likely risk of publishing your genome, which is revelation of false paternity. If you would have turned out half only Ashkenazi, your mother would have had some explaining to do 1.

    The more serious issue was already touched by Lisa. What about your children? As I have thought about this issue, I see that my parents imposed my genome on me without asking my opinion, hence I am free to do what ever I like with it (including publishing it). But since I’ve done the same thing to my children I have an obligation to protect their genomes from any abuse. (The issue I haven’t thought out is whether publishing someones genome is/enables abuse)

    1. http://dienekes.blogspot.com/2010/10/running-euro-dna-calc-on.html

  • What I’d give to be a fly on the wall at the Vorhaus dinner table.

  • Daniel MacArthur

    Hi Marnie,

    Sorry about the delayed reply.

    Because of scientific and public bias, as well as public misunderstanding, the your freely available genomic data is likely to be both misused and misinterpreted.

    Obviously we’ll do our best to avoid this by providing clear explanations on the site, but it’s impossible to prevent all possible forms of misunderstanding or intentional abuse. We hope that as such examples of such problems arise we’ll be able to publicly correct them here – hopefully reducing the risks for future public genomes.

    Hi Kimmo,

    As my wife and I are both involved in this project, obviously we’ve thought pretty hard about the potential implications for our son. While we obviously have no intention of making his genome public until he’s old enough to give informed consent, the presence of both our genomes obviously has some implications for him.

    Working through the probabilities is reassuring (unless both of us are homozygous for something, no-one can predict with certainty what his genome would contain). On balance, we suspect that this decision is likely to do him far less harm than other decisions we make as parents over the next couple of decades.

    Still, there are no guarantees here: uncertainty is the price we pay for doing something before all of the potential implications have been fully worked out.


    You ask some excellent questions. I’d like to see if other group members want to weigh in before I do – but these are very good discussions to have.

  • I am a newly-minted member of PGP 1k who is about to start a research project about the social influences and personal attitudes that impact informed consent for WGS. This was perfectly timed for me, and I found it thoughtful and very personally helpful. Hopefully you’ve set the stage for more sharing to come. Thanks.

  • First, apologies for the delayed reply. There are some great, thoughtful comments and questions here, and I’ll do my best to reply in order.

    Understanding Genomic Data (@Marnie): As you point out, one challenge for Genomes Unzipped, the Personal Genome Project and other projects that facilitate the public sharing of genomic data is to make sure that participants are able to understand their own data. That is one reason why Genomes Unzipped is not, for now, a repository for anybody and everybody in possession of genomic data to share that data publicly. It also illustrates the importance of a robust informed consent process.

    The issue you raise is also one of the reasons I chose to participate in Genomes Unzipped. While I can’t speak for my collaborators, one of my hopes is that by exploring our own genomic data, and by opening it up to intentional or accidental misuse or misinterpretation by others, we can help increase the ability of others to understand their own genomic data.

    Because we cannot – and should not – restrict genomic data to individuals with scientific training, we need to explore other models that facilitate a high level of understanding. I believe Genomes Unzipped can serve as a resource in that regard so that more and more individuals will be prepared when they encounter their own genomic data.

    Familial Informed Consent (@Lisa): It’s a tough question indeed, where to draw the “consent cut off.”

    In many ways my parents and siblings represent the easiest case. They are all living, above the age of legal consent and their whereabouts are known to me. Other situations – including unborn children, relatives who are incapable of providing consent due to their age or mental fitness, relatives who can not be located, etc. – are much more difficult.

    In every case, however, I think the fundamental issue is the same: what do you do when multiple individuals can assert a legitimate interest in the same information (in this case, genomic data)? How do you balance a prospective participant’s right to control access to her genomic data with her family members’ rights to limit access to that same data?

    For the prospective participant, I think the right shapes up as a form of freedom of speech, although I don’t believe that the publication of genomic data has ever been asserted as such in a court. For the family members, it is more like a privacy right.

    There are plenty of other complicating factors as well. The calculus might change if there was a demonstrated harm associated with the publication of genomic data, depending on the nature and magnitude of that harm. Similarly, there could be a public interest in genomic data (as there is now with certain other health data, such as in the case of infectious disease or other public health reporting).

    How are we to balance these competing interests in genomic data?

    My own thought is that, unless we consider genomic data to be of a qualitatively different nature than other forms of personal, private data – which I do not, at least today –the individual’s right to publish ought to trump the relatives’ rights to privacy.

    Clearly not everyone agrees with that calculus, and it’s a determination that I think depends at least to a degree on context. @Lisa and others, what do you think? The lack of consensus means that this issue will continue to crop up, and it may even be the type of scenario that will lead to some very interesting litigation somewhere down the road.

    With respect to Genomes Unzipped, I was able to largely avoid a difficult choice by taking the time to talk through this issue with my family. We talked at length about Genomes Unzipped, and particularly what publishing my genomic data could mean for my family members, including the risks it entails. I’ll be writing about that discussion in more specific detail in a future post but, as is clear from my involvement in Genomes Unzipped, they all ultimately supported my decision.

    Finally, as for the portion of my family incapable of giving consent (i.e., my unborn children), an issue also raised by @Kimmo, I’m on the same page as @Daniel. While the publication of my own genomic data has some implications for my future children, I believe that any risks are too far remote in both time and probability to require me to refrain from participating in a project like Genomes Unzipped.

    There is no doubt that the issue of familial informed consent for public genomics projects (and, to a lesser extent, for non-public genomic projects as well) is an issue that we have just begun to grapple with.

    Thank You Notes:

    @Misha: you know you’re always invited.

    @Leila: Welcome to the PGP-1K! I’m glad that you found this discussion helpful. Hearing that helps to affirm my own decision to participate in Genomes Unzipped.

Comments are currently closed.

Page optimized by WP Minify WordPress Plugin