I work in genomics. My lab, many of my friends and acquaintances work in genomics or had to dabble with it for at least one of their projects. You would think we would all be fascinated by this stuff, getting to know your own DNA and everything. Yet I don't know a single person who would even consider using 23andMe's service.
To me this is all an elaborate scam. Why on earth would you pay them to give them your data?! At least with Google et al. we know that if you are not paying for the product, you are the product. With this company you end up $100 (or whatever) short and you surrender your data for it to be sold to advertisers, insurance companies and whatnot. And not just stupid things like your Amazon shopping history or your latest Tinder conversations (which are in themselves pretty damn intimate if you ask me) but the most intimate thing of all. I actually disagree with behavioral genetics/sociobiology fundamentalists who think there's little more to the self than one's DNA, but just because one cannot make much out of it doesn't mean it can't be misused. If Gattaca implications sound scary, imagine what would happen if the decision makers had no idea what they were doing in the first place.
Because that's where we are at this point: just because X variant is associated with Y% more incidence of whatever disease does not mean extra care should be taken regarding risk factors, insurance policies, etc. The correlations are interesting when combined with other data but most of the time we have no idea what's going on and what it is that makes variant X cause disease Y, if it does at all. Doesn't mean it won't be misused.
Please people, don't pay to get your data swindled out of you. Stop with this weird fascination with your DNA, and stop trying to look for an answer when are barely asking the questions.
Allow me to offer a different perspective: I've used 23andme and it was one of the best decisions ever.
Not so much for the Ancestry part, which was also interesting and explained my above-average cold tolerance, but for the health stuff.
Granted, being from Europe they don't directly offer health services, but you can get the raw data and upload it to various sites for interpretation -- ranging from really accessible to follow-the-rabbit-hole style.
I've accepted from the start that everything is uncertain, or just a possibility, but it was still very useful, because I knew what to look for and thus able to validate easily if true or not.
Some things that were in the reports I already knew, some I suspected, some I had no idea and would have never guessed -- this last group had the most impact.
Some examples: allergies, motion sickness, needing above average amounts of vitamin C.
All very easy to test, but with an amazing quality of life improvement gained as a result of just changing some simple things.
So yes, a US company has my personal DNA data, maybe they will take care of my privacy, maybe not, but the quality of life gains were worth it, for me.
What other alternative would people like me have, from countries where there is no access to good doctors, of finding stuff like that, had there not been 23andme? Especially since I didn't even know what I was looking for.
That's a very good and level headed response to the parent. I agree with both perspectives. I think the crucial point is that you seem to have thought through the implications of deciding for or against using 23andme and after weighing the outcomes, you made an educated decision. Perhaps the decision was right (fingers crossed), maybe it was wrong, but the important point is that you thought about it.
I took parent's post less as a blanket warning against doing DNA tests, and more of a caution to consider the implications.
I think there is value in services like 23andme, and even if there are risks that the information will be misused, the risks are worth it. I don't mean the risks associated with case of a specific person, but the research and services associated with DNA sequencing in general.
Maybe 23andme is not the way forward, and eventually some other (better) services will arise, but we (humanity) are in something of an uncharted territory here, so there is need for a bit of trial and error. Yes there be dragons, and occasionally we'll have our eyebrows singed, but advances in medicine and biology (as your case highlights) are vast.
What sites did you use ?
I used promethease site, but it was way too much information, and the filters felt complicated.
Is there any other site that makes a better job filtering what is more important for you specifically ?
Any recommendation ?
To add to this, though genetic information may be protected against defining a pre-existing medical condition for insurance purposes today, there's zero guarantee that that law protecting you will still exist in the future, or extend to your descendants (which can be assumed to carry roughly half of your genetic material).
Once you've given this data away without any guarantee that it'll be destroyed at the end (if such a thing is even possible), you can't take it back, and you could potentially be screwing over those who didn't make that choice.
Thus comment and the parent are spot-on. The main protection against the adversarial use of genetic databin the US, GINA, is currently facing several challenges, mostly through the backdoor of opt-in employee wellness plans.
Additionally, while the SNP- format generic data provided by 23andMe is of marginal medical use to the average consumer, it is very valuable to insurance companies and others constructing population-level actuarial models, where a very tiny increased probability of developing a condition is enough to justify increased rates.
And finally, as these relative-discovery stories suggest, there really is no such thing as anonymized genetic data above a panel of a few SNPs. This data is part of the inherence class of factors in multifactor authentication, and can even be derived from pooled anonymized data by a motivated party.
What a group of friends has been doing is, they ordered +10 kits over a year shipped to the very same person, then they each of them spit and register the kit with a disposable email address. The info about the receiver is watered down (if he did one himself, which he didn't) and the rest remain reasonably anonymous as long as they keep using Tor/VPNs.
Moreover, these tests are forbidden in France, so these shipments are being sent to Monaco/Italy (1 hour drive from here), adding extra levels of law/tracking indirection.
Keeping your name and location anonymous doesn't matter. Once a company obtains a family member's DNA information through any other source, they will see a familial correlation, use other records to realize that family member has a brother/sister/etc and might now attribute that once-anonymous DNA record to you.
That's the scary thing about these companies: you might never even use the service, but once a relative does, the company now has information about you.
If insurance companies wanted genetic testing data they wouldn't bother trying to do a messy data join with a third-party database, they'd just make it a requirement to provide a swab/saliva sample for testing when you signed up.
Boiled frog fallacy. They won't be able to get away with it now, but as time goes on they'll start adding it as a voluntary option for a small discount, eventually ceasing to offer service otherwise.
Auto insurance is already well into the process of doing the same thing with regard to diagnostics data being transmitted back home from Snapshot devices/driver's aids.
I paid for 23andMe. I did it mostly out of curiosity, to see if there was anything in their genealogy report that would surprise me (there wasn't). I was also interested in their medical risk reports, although I admit to having very little knowledge about how reliable or extensive those reports are.
The thing is, I actually agree with all your concerns about privacy, but I guess I just disagree about the magnitude of the concerns. Of course there are risks of being targeted for ads, or treated differently by health insurance providers. I can't really justify my stance, but I guess I'm just less worried about the possible downside than I was interested about the possible upside.
The thing is, though, if the world becomes as much of a dystopia where the people who have their data on file at 23andme causes some sort of major havoc in their lives, why do you think it will stop there?
Either the world will turn into a dystopia or it won't. If it does, not giving out your DNA sequencing information is, in all likelihood, not going to allow you to avoid getting caught up in it.
The thing about a dystopia is that you can't control everybody or else you will incite a revolution. There is actually a balance that you need to maintain. Privilege is one of the main ways of maintaining slavery. For example, if you are 1 person and you want to control 10 slaves, it is not really possible -- they will gang up on you. However, if you make one of the 10 slaves the "head slave", then they have an incentive to maintain the status quo. If you then make it clear to the other slaves that they can be upgraded to "head slave" as you expand your slave empire, then they will also have incentive to maintain the status quo. Your goal as a slave keeper is simply to make the reward/liability comparison work out ever so slightly on the reward side. You want people thinking, "Well, if I work hard, then I might be made head slave. And my slavery is not so bad. If I try to run away or rebel, then I might be beaten -- and I don't even know if I can survive if I run away".
No matter what dystopian world you live in, you will always be given the illusions that life is better as it is than it would be if you were free. They aren't going to throw you to the wolves completely. Instead, they will always present you with a dilemma: "As long as you only do X, you'll be fine, so there is no need to fight against it". In terms of privacy, there always has to be the question, "Why do you need so much privacy anyway?" If you answer that question, then you will break your balance (see recent Facebook "scandals" -- nothing changed except the general understanding of why people wanted privacy).
The advice to avoid problems is a pragmatic one. "If you do X, you'll be fine" is funny because people will say that and then forget to do X. "Well, I'm sure I'll be fine anyway". If you are careful to do X, then you can avoid many of the downsides. You can use that time to break free from your slavery. Umm... Or not... (usually people choose the "or not" option).
I travel a fair bit and stay at the hotels here and there. I keep wondering, how long before hotels start sampling your personal belongings (toothbrush, comb etc) for gen data. Maybe that already happening? From there, how long before full lists of these data {genetic sample; full name and other credentials} are on sale on some special market?
Call me paranoid... is that sort of activity even regulated? Isn't this like a public image for photography?
Well, if we're being paranoid they probably don't need to touch your personal belongings. They just have to have a look at the bed sheets after you've slept on them.
...or to get your DNA they can work it out from the "deposited" DNA of your close relatives. There will be no hiding! :P
Sure! What I specifically meant (sorry, was unclear) is, how long before such activity becomes lucrative enough for hotels to routinely perform? ("hotels" == entities in a completely unrelated lines of business) I.e., there is stable market for these data and the procedure isn't very costly in recurring expenses (no costly extraction from "noisy" support).
if you were born in the state of california after 1966 then the government did collect and is retaining your dna (newborn blood specimen) and it is a law.. already
According to "An Association Between the Kinship and Fertility of Human Couple" (http://science.sciencemag.org/content/319/5864/813.long), third (or fourth) cousins have optimal reproductive success. Reproductive success is not just the number of children (because 1st or 2nd cousins couples tend to have more children but with associated health and/or reproductive problems), but the whole descendance.
This is not a big deal. In fact likely better fertility for them: a 2008 deCODE study results show that couples related at the level of third cousins have the greatest number of offspring, with the greatest reproductive success observed for couples related at the level of third and fourth cousins. [1]
I would be enormously surprised if genetic factors were the primary driver in this. I would think cultural norms and socio-economic factors would be the greatest predictors of the number of offsprings.
I could easily envision that those same factors could contribute to a rise in intermarriages between 3rd and 4th cousins. For example, in small communities like "hill people" in the Appalachians, or the Hasidic community in NY.
The sociological effects of DNA matching sites are going to be interesting to watch. I personally found out a year or so ago that my dad wasn't my biological father. At the age of 45, that's an, um, interesting secret to suddenly discover.
Now, imagine what happens in a society with strict laws/rules/mores about this sort of thing? Well, we won't have to imagine for long. I'm quite sure my experience is a lot more common than most people realize, and I think much of the world isn't ready for it.
> I'm quite sure my experience is a lot more common than most people realize, and I think much of the world isn't ready for it.
Nonpaternity is quite rare overall, and still unusual even when the putative father has so little paternity confidence that he demands a paternity test.
I wonder if there how those low numbers have changed over time. I ask from personal history.
My grandfather was away in WWII when one of my aunts was conceived. As my grandmother put it, she was lonely.
All of the adults at that time knew, but kept it a secret. None of the kids knew until that aunt was about 45 and diagnosed with a cancer ... with a genetic susceptibility passed down though the male line. My grandparents figured it was time to reveal that secret.
I therefore conjecture that the numbers in the US were higher during WWII.
On the other side of the coin, I know someone who recently found his biological parents at age 50 through one of these places. Turns out both he and his biological parents had tried to find each other in the past but were stymied legally.
This kind of thing exposes secrets. Some of those secrets could be uncomfortable (I'm truly sorry you have to deal with what you discovered!), but some can be good. It'll be interesting to see how society going forward reacts to biological ancestry secrets not being able to be swept under the rug so easily.
It's touched upon in the article, but in reality they're probably not third cousins.
23andme gives relationship estimates for people based on percentage of dna shared, these estimates are based general population data. However they're not accurate for endogamous
population like Ashkenazi Jews.
Any two Ashkenazi Jews who get married are likely to show up as cousins to each other, simply because there's so much shared DNA within that community. This isn't an exceptional case, but rather the norm for such populations.
You share (about) 1/128 of your genes with a third cousin. The chance of any (particular) recessive trait being passed on from both parents as a result of the inbreeding is a whopping 1 in 65,536.
I'm not sure what you computed. The interesting question is to pick a recessive trait from one of the parents and ask for the chance of it being passed on. The probability that it's being passed on from the parent who has it is 1/2. The probability that it's being passed on from the other parent is (1/2)*(1/128). The combined probability is 1/512, not 1/65,536.
I'm not an expert on this, so perhaps you could tell me if a family history of cousin marriage increases the likelihood of the recessive trait being passed by both parents. I mean, if for the last 10 generations a family has been marrying cousins to each other and introducing little variation from people outside the family, the odds are increased, right? I'd imagine cultures with a history of cousin marriage, such as the Pakistani community in the UK, are much more prone to this type of problem than when cousins marry, but none of their forbears married a cousin.
Giving birth after 35 (in this case 37) is statistically more risky than offspring from first cousin's. A third cousin is much more distantly related than a first cousin, your first cousin has a pair of the same grandparents as you, your third cousin shares a pair of your Great-great-grandparents.
A relative of mine who is a doctor mentioned to me that the opposite is a common problem when children volunteer to donate an organ to a parent. They must test for DNA compatibility which often yields bad surprises. Apparently it is gradual. The DNA of the first child almost always matches, the second a little less, the third is where the odds reduce significantly.
I'm curious of the legitimacy of this story, given one of the comments at the bottom:
> Don’t believe the results. I’m retired FDA and investigated this company since its inception. 23and me swabs go to labcorp in NC. They aren’t a lab. They take the results and issue what is tantamount to a horoscope based on your demographic information. Save your $100 or whatever they are charging now.
This is a ridiculous claim. It might be less accurate than other options (although I doubt that), and misses some genetic features (repeats, insertions, translocations or deletions), but there's a huge distance between that and "a horoscope based on your demographic information".
Third cousins sounds like a non-issue to me, and it sounds like the author and spouse generally agree. I've never knowingly met any second cousins or any other descendants of my great-grandparents (who are not also in my grandparents' lineage).
I wonder if there are any estimates on how common unaware third-cousin marriages are.
It’s interesting how much that can vary. I’ve met every one of my second cousins on 3 out of 4 family branches (one grandparent was estranged from their family). Those who live nearby I’m closer with than first cousins who live further away.
My and my siblings’ children will also know some of their third cousins, and see them at least once every few years at holidays.
I’ve met a couple third cousins, and am fairly certain with any others (aside from that one branch of the family), we’d discover our relationship well before being together long enough to consider marriage.
That said, I wonder to what degree physical proximity plays a role. If the second cousins didn’t for the most part live in the same metro area as the entire rest of the family, we’d probably see them even more rarely than the set of first cousins who live a 3-hour flight away - which is to say, roughly once a decade.
It is fascinating how people grow up with such varying connections with their extended family. Even growing up I would only see my grandparents, aunts and uncles, and first cousins once or twice a year at major holidays.
Most years there would be a nearby annual family reunion of my mother’s mother’s parents’ lineage. I remember it mostly as a lot of older people I didn’t know, and none of them lived near me (the venue was a 2-3 hour drive from where I grew up, and I don’t know how far other people drove.)
But in my high school class there were definitely lots of people who seemed to be cousins with each other, and had massive family reunions in town. In high school my friends always joked about these extended family events as a hassle, but part of me thinks that perhaps I missed out on something there.
It’s interesting to think about the factors that influence a child’s experience with extended family. I grew up only two hours from where my mother grew up, while my father had moved quite far before meeting my mother. My mother had two siblings and my father had one, which is perhaps on the low side for America in that time (my parents were born in the 1950s).
Another interesting thing is that I have never met anyone outside my immediate family with my last name. My father only had one sister, who was married, so those cousins have a different last name. My father’s father died young (long before my parents met), and my grandmother remarried, so she had a different surname for my entire life.
I grew up in a very small town in a rural area, and I would have expected this to be the sort of place to be where people don’t move far from home and where extended families stay close together. But even my extended family in the same US state (but a couple hours’ drive away) was never super close to me. I barely escaped high school before social media went mainstream, so I wonder if I would have forged closer relationships with my extended family with the help of social media.
Asked about the biggest mistake her startup made in the founding phase, she answered that " I think that we were overly optimistic about the state of scientific literacy in this country. "
And that is true in general. I live in Germany, I am a 23andMe customer, and from my results I learned quite a bit about myself. Locally, whenever I tried to talk about this genetic testing service, I have never had a deeper conversation about this subject matter, even with highly educated folks. It's similar to non-conversations about email encryption and IT security many of you readers might have had.
The author alluded to the link that her second child is autistic and the relatedness with her spouse.
But what's more glaring is that her second conception occurred when she was 37 and her husband 39. There are studies that showed the increased incidence of child autism conceived from older fathers.
Two thoughts came rushing to my mind after reading this.
Disclaimer: I have not researched the scientific background of these thoughts. They are mere thoughts.
It made me uncomfortable to read about the autistic child at the end of the story. Was Autism more probable because of the closer genetic relationship of parents?
I was raised in southern India, and some of our families have a fascinating astrological tradition about not marrying someone from same "Gothram".
Is it possible that our ancestors knew that certain genetic combinations were unhealthy/undesirable even without a scientific background and adopted these rules for the society?
I am from India(South) and also coming from a family where there are quite a lot of people in the medical field.
Yes it was known not to marry inside the family by our ancestors. I know several people who are suffering from various birth defects due to marrying very close and inside the family and yes it is told by the doctors.
Gotra is completely screwed up now. A system has to be taken seriously. If a person doesn't know his Gotra then he is just arbitrarily assigned one which becomes dangerous when done in huge volumes.
Also in India there are caste communities which are smaller and when they have to marry inside their families problems start cropping up.
There is a huge demand for home nurses and speech and hearing specialists in the Gulf region. This is because of consanguineous marriages very prevalent there. There are a lot of children born with mental or speech disabilities.
Nearly every culture has taboos on incest. https://en.wikipedia.org/wiki/Incest comments "sexual relations with a first-degree relative (such as a parent or sibling) are almost universally forbidden".
As the Wikipedia page I linked to further states: "cultural anthropologists have noted that inbreeding avoidance cannot form the sole basis for the incest taboo because the boundaries of the incest prohibition vary widely between cultures, and not necessarily in ways that maximize the avoidance of inbreeding."
The further refinement of your question might be, how do the unique aspects of the astrological tradition affect inbreeding?
In order to test your hypothesis you would need to define gotras. I will quote from the Wikipedia page you linked to:
> the definition of gotra as descending from eight sages and then branching out to several families was thrown out by the Bombay High Court. The court called the idea of Brahmin families descending from an unbroken line of common ancestors as indicated by the names of their respective gotras "impossible to accept." The court consulted relevant Hindu texts and stressed the need for Hindu society and law to keep up with the times, emphasising that notions of good social behaviour and the general ideology of the Hindu society had changed. The court also said that the mass of material in the Hindu texts is so vast and so full of contradictions that it is a near-impossible task to reduce it to order and coherence.
Sadly not all societies/cultures have a taboo against incest:
"Couples who are getting married should be forced to have a DNA test first to ensure they are not cousins amid growing concern about incest within Pakistani communities, Britain's first Asian peer has claimed."
>Third-degree relatives (such as half-aunt, half-nephew, first cousin) on average share 12.5% genes, and sexual relations between them are viewed differently in various cultures, from being discouraged to being socially acceptable.
Relationships between cousins isn't always considered to be incest. It's more recent in the last 50-60 years, especially in USA, that cousin marriage is frowned upon based on misunderstanding the risk.
The other issue there is people who are related by multiple lineages. I know a family where the brothers married 2 sisters. Technically, their kids are cousins, but genetically they are siblings. Do enough of that and your genetic diversity of a community goes down to the point recessive diseases start showing up.
Several commenters in this discussion have taken as read that the couple are third cousins. Reading the whole article, one finds buried at the bottom the fact that this turned out to be unproven, because the genetic tests are based upon general assumptions that are not true for their segment of the population.
We're all related, really. It just depends to what degree you want to consider "related". I'm related to my cat, to you, and even to the tree outside my office if I go back enough generations.
Considering how this and my comments have been flagged we hit right on the mark. You can't casually use sexist remarks towards women, but apparently not only is it OK to use them towards men, but mentioning it as negative results into flagged post.
To me this is all an elaborate scam. Why on earth would you pay them to give them your data?! At least with Google et al. we know that if you are not paying for the product, you are the product. With this company you end up $100 (or whatever) short and you surrender your data for it to be sold to advertisers, insurance companies and whatnot. And not just stupid things like your Amazon shopping history or your latest Tinder conversations (which are in themselves pretty damn intimate if you ask me) but the most intimate thing of all. I actually disagree with behavioral genetics/sociobiology fundamentalists who think there's little more to the self than one's DNA, but just because one cannot make much out of it doesn't mean it can't be misused. If Gattaca implications sound scary, imagine what would happen if the decision makers had no idea what they were doing in the first place.
Because that's where we are at this point: just because X variant is associated with Y% more incidence of whatever disease does not mean extra care should be taken regarding risk factors, insurance policies, etc. The correlations are interesting when combined with other data but most of the time we have no idea what's going on and what it is that makes variant X cause disease Y, if it does at all. Doesn't mean it won't be misused.
Please people, don't pay to get your data swindled out of you. Stop with this weird fascination with your DNA, and stop trying to look for an answer when are barely asking the questions.