23 And Me
A month or two ago our daughter Jessica asked us to get 23andMe kits and get our genes sequenced. Jessica wants to add some information from our results to hers so she can better understand her own genetic makeup. So the Gotham Gal and I did it a few weeks ago and we are now awaiting the results.
The 23andMe process requires that you set up a profile on their web service and answer some questions about yourself before submitting the saliva sample for processing. If you wish, you can give them a full health care profile which is almost exactly identical to the interview a new doctor will give you on your first visit (do you/did you smoke, do you/did you drink, are your parents alive, are they healthy, are you allergic to this and that, etc, etc). I went through the entire process and shared all of my medical history with 23andMe in great detail.
So now 23andMe will have a full medical history/profile of me plus my genetic makeup. That’s pretty cool.
I have some suggestions for them as a user.
First, I wish that I could connect my profile on 23andMe with Jessica’s profile on 23andMe as her father. And I wish that I could connect my profile to the Gotham Gal as her husband. In summary, I would like for us to build a family tree on their service so that the work Jessica wants to do can be done automatically by their software. That seems like an obvious thing to do in a service like 23andMe, but I looked around pretty hard and could not figure out how to do that.
Second, I would like to be able to authorize third party services (starting with my doctor’s Patient Fusion service) to access all of this genetic and medical information I have stored at 23andMe. I found the 23andMe API so that’s a good thing. But I could not find a marketplace of third party apps that connect to 23andMe via its API. That would be really useful.
Another thing I would like to be able to do is sync my 23andMe data with my mom’s genealogy data that she keeps on Ancestry.com. I believe she has some basic genetic information there on herself. That would inform my profile in the same way that my data informs Jessica’s profile.
The bottom line for me is that this data (genetic information and medical history) is really powerful stuff. It should not be held in silos. Users should have the power to move it around, connect it up, and share it with their medical providers, family members, and others who can benefit from this data.
This is a big part of our thesis in health care. Users are starting to get control of their own medical data and decide how it will be used and by whom. That will lead to better outcomes for everyone. I am very excited by the potential of what can happen when this is really happening at scale. And that feels like it is right around the corner now.
If you want to get a 23andMe kit, you can do that here.
Fred, is privacy of your genetic makeup something you think about?I could see a breach of such data being much worse in the long term than credit card or social security data.I’m not sure how I feel about third party apps being able to access that stuff.
i don’t care. i’d publish it on this blog if i could. why should i care?
That medical report from my college days and subsequent symptoms is a tad embarrassing. I dunno, maybe that’s why?
Maybe you wouldn’t care personally. But I can see a situation where there is a data breach and millions of peoples genetic makeup becomes available to outsiders.Less than ethical HR people could cross reference a candidate for genetic markers for various personality traits.Lenders could correlate their internal data with genetic data and come up with a genetic profile who is less like to pay back and loan, and make decisions on that.And then there is Gattaca…
I’m with Fred as I simply don’t care about this–and honestly about much of the things privacy advocates harp on.But for the Gattaca reference–I thank you and gonna cue it up to rewatch.
Here is a little Gattaca backstory, courtesy of 23andme patent applications.”Taken out of “patentese,” what 23andMe is claiming is a method by which prospective donors of ova and/or sperm may be selected so as to increase the likelihood of producing a human baby with characteristics desired by the prospective parents, the selection being based on a computerized comparison of the genotypic data of the egg provider with that of the sperm provider. The phenotypic characteristics that may be on the users’ (e.g., parents’) “shopping list” can include both disease-related and non–disease-related traits, such as height, eye color, muscle development, personality characteristics, and risks of developing age-related macular degeneration or certain types of cancer.”http://www.nature.com/gim/j…
Nice detective work.
Read my comments below it is shocking how right some moviemakers predict the future. The robot predicted by Ailen’s in 1979 to exist in 2022 is here. I’m sure other movie buffs will have better examples than this, but Gattaca is set in the same timeframe.
I don’t think it’s shocking at all. We use sci-fi as our map of future technology, consciously or not. We invent the future we invision, for the most part.
Yep, that’s why we’re living a re-mix of 1984 and Brave New World.
erm…parts of this happen already….
Lets harp on this proverbial harp:…lets just say that your publicly “gettable” genetic info placed you in a category that was more susceptible to Alzheimers and that a potential employer that requires high cognitive function happens to know this … this would fundamentally alter their decision to hire you in a manner that was not possible before.Apply this rationale to the bank that gave you a house loan or the insurance company that plans to overcharge you based on the likelihood of you getting an expensive disease.
honestly of no concern to me. you need to choose the things to worry about in life. there are plenty but that’s not one.
Insurance is to socialize risk. If you are virtually certain to get a disease, then that sucks, but it is not the obligation of the insurance company.On the other hand, that is a reason for socialized healthcare.
While I don’t care about this in particular either I do care about limiting what I will call “the attack surface”.The point being that you have to weigh the gain of publishing 1 blog post with what the possible negative consequences might be. Unfortunately that is an unknown. So perhaps it’s better to err on the safe side given the nominal upside.
i seriously don’t care that much about most of it.Some of course.I lived 4 blocks from where the towers were. If there is a degree of my personal privacy that needs to be compromised to keep events like that from happening, so be it.
100% with you on that one (and I only saw it on TV).I think part of the problem is the juvenile perception of some people whereby they think that if the government has access to certain information they will, for sure, act on that information. That is not the case. There aren’t and there will never be enough resources to even get everyone to pay taxes correctly (which would be trivial to do) let alone the long tail of crimes that people are apparently hiding from.
100% with you on this, Arnold. Often wondered if diehard privacy advocates lived in NYC and experienced 9/11 firsthand they’d have a different POV. One could say that’s an emotional response and I’d reply that emotion (and, of course, saving lives) is a big part of my reasoning. I know of a few families who’ve been able to move on, but many haven’t. I know one family who lost a son and both his parents died w/in a year from stress and heartache.
Yup, its a tough reality for many.I have a few friends who left the city, and still to this day feel the loss of being away even though some family members just couldn’t deal with the after effects and stay here.
I didn’t live there in 9/11 but I used to live at Liberty and Nassau.
This is the sum of my feelings towards the experience.I republish the same post every year as I simply can’t express it better and I feel the same still after all this time.September 11th…stopping to remember http://awe.sm/jPMzB
Interesting, actually I am going to do the same re rewatching Gattaca.I have gotten into learning about genetics a bit over the past several years. When I first saw Gattaca in the late 90’s, I knew absolutely nothing about genetics (it was before the genome was sequenced as well) and I have not watched the movie since. Will see if my reaction to the movie turns out to be different when I watch it now.In other news – Very exciting news coming out of LIGO over the past hour confirming what Einstein predicted a 100 years back ! (And with where genetics can take us, maybe in the future people will see more of their predictions proven in their lifetimes).Is this a great time or what !!!
Couldn’t imagine a better time to be alive and productive.
Just don’t leave anything for the swim back…
You know there are new approaches now to encrypt data not only during access, but also when you store it and the app uses it, so even if a breach happens, the hackers get encrypted garbage. At least, that’s the modern way to write software applications.
Sure. But I think it’s naive to think that valuable data won’t get out there some how, some way.It could also be ‘shared’ with the government or other corporations voluntarily. Perhaps they have less rigorous standards.Data breaches always happen at the weakest link.
I agree.You need to be somewhat at peace that this is going to happen.
Couldn’t the same argument be applied to hospitals, doctor offices, etc that are all converting to EMR? I’m sure hacks would be way more likely here than high-tech companies that are highly proactive and more capable at protecting data.
happens all the time, but so far it seems to be mostly for identity thefthttp://www.forbes.com/sites…Healthcare’s “wall of shame” for 2015 officially ends tonight at midnight. It’s not really a “wall,” it’s just a website, but it’s the online mechanism for the Office of Civil Rights (OCR) under Health and Human Services to publish data breaches as reported to them and required by HIPAA. The numbers this year are just staggering.According to OCR, there were 253 healthcare breaches that affected 500 individuals or more with a combined loss of over 112 million records.The top 10 data breaches alone accounted for just over 111 million records that were lost, stolen or inappropriately disclosed.The top six breaches affected at least 1 million individuals–and four of the six were Blue Cross Blue Shield organizations
Having worked at 23andMe and then in enterprise healthcare IT, I can say confidently that I would trust the former way more than the latter. The level of IT security in most healthcare provider organizations is pitiful.
There have been many high profile security breaches at high-tech companies in the last several years, many of them household names in the tech sector. They definitely are more capable, but some do not give it the importance it needs.
There are weak points that can always be breached. There is a key to the data that is available it’s not locked up like nuclear secrets either.
And that’s exactly how we wrote it. While no system is foolproof, security was a huge consideration from day one when building the 23andMe backend infrastructure.
Would imagine you could post data for people to aggregate, reference, etc – without having to disclose your identity. Identity preference settings should be key.
The Gattaca reference is actually pretty compelling to me.
That was the first thing I thought and it scared me how accurate some moviemakers have been. Today I am at a robotics conference. http://www.wearablerobtics.com.You can look there are tons of exoskeletons. One person showed the one Sigourney Weaver wore in Aliens produced in 1979. The movie was set in 2022, we got there 5 years early.
Of course, engineers are all sci-fi fans 😉
Actually I am not. Does that nullify my engineer card?
From the site’s TOS:”You should be careful about sharing your Genetic Information with others. Currently, very few businesses or insurance companies request genetic information, but this could change in the future. While the Genetic Information Nondiscrimination Act was signed into law in the United States in 2008, its protection against discrimination by employers and health insurance companies for employment and coverage issues has not been clearly established. In addition, GINA does not cover life or disability insurance providers.”
There is a theory that pieces of the Breast Cancer supercluster is related to intelligence. I would avoid Gattaca and HRing genes.If I am not mistaken, I also believe if there was a data breach, US law prevents discrimination on that basis
You can download your full raw data file here, it’s a small enough file to publish easily: https://www.23andme.com/you…
Ooh. When I get my data I may do that
Hah! The point of data that people have now have is that if you don’t publish it’s because you have something to hide. In all seriousness though I think it’s a mistake to do this.
To add to my other comment “mistake”. For one thing you will encourage others to do the same and for those people it might matter and have an impact (employment, insurance or some future thing that nobody has thought of). In other words things, by virtue of who you are and your financial situation don’t matter.
Fred, out of curiosity, would you be willing to publish all your medical records on this blog?
Surely there’s a difference between the two? Your DNA isn’t exactly private data (you leave it everywhere). Specific acts of health are more easily kept private, and people might be more interested in that.
My DNA is not private as I leave it everywhere? Have you come across any of it recently?
Do you consider your fingerprints to be private data?And yes, everytime you throw out garbage, you probably have left some DNA of yours in it. Any time you shed a hair in public, you left your DNA in a public space.
Yes, I consider my fingerprints to be private data, which is why I would be very upset if I happened to come across a scanned copy of them, next to my name and face, online.
I think your definition of private is an odd one. How can something that you leave literally everywhere you go be private?
I’ll try and explain this. For starters, I don’t leave my fingerprints everywhere. Just in the few places I habituate, which is probably less than 1 millionth of the surface area of this planet. In addition, even if I left a fingerprint on a random taxi door, and you came across said taxi, you still can not link that one out of the hundreds of fingerprints on the taxi to me, unless you knew who I was and that I had been in that taxi, and had the fingerprints of all the other hundreds of people that had touched that taxi to eliminate theirs first.It’s like saying your credit card numbers are not private because you hand the card to 10 strangers each day.Private does not mean anonymous.
I think you’re confusing the idea of securing data with the idea of keeping data private.
No I am not. I am simply stating that private data is information whose use/dissemination you can be reasonably expected to have some degree of control over.
That’s precisely my point. You leave your DNA everywhere. It is not hard to get someone’s DNA, because (short of wearing a frogman’s wetsuit everywhere) you shed it all the time.How often you leave fingerprints isn’t the consideration of whether they are private data. It is personal data, but that doesn’t necessarily make it private.
The fact that my physical DNA can be gathered by someone close to me does not make a digital copy of my genetic code public information, and more importantly, it is extremely difficult to get someone’s DNA. You make it sound like it is as easy as downloading a Coldplay MP3 – it is not. I challenge you (specifically) to go find, and sequence a copy of my youngest niece’s DNA, and let’s see how long it takes you to do that. The fact that I leave fingerprints all over the place does not give Apple the right to keep all their customers’ Touch ID fingerprint scans as un-encrypted JPEGs and publish them on wikileaks.
Well, if you have used 23andMe, then you’ll know that the information they produce *is* downloadable and sharable in a raw form. Size of the data isn’t an issue. People regularly download GB and TB of data in a shared fashion.Again, it doesn’t matter if someone is targeting specifically you or not, once you drop a piece of hair or leave some saliva or blood in public somewhere, then you have just publicized your DNA. Full stop.
I don’t understand where you are going with this. Even 23andMe clearly state that your genetic data counts as sensitive personal information and is covered by their privacy statement.
How terrific for them.
Fred – just curious – you have an opinion on editing DNA? Looks like it’s only a matter of time before this technology really works. I can see both sides of the coin on this super ethical dilemma. On one hand – you can rewrite one’s code to live a healthy life and avoid conditions like cystic fibrosis. However, I’m sure this technology won’t be cheap and could eventually be a tool for the rich to ensure genetically superior beings / “designer babies”. I know this isn’t really a new question – just wondering where you stand. Maybe it’s a scenario where you need rules and regulations to draw the line?
if someone has your DNA, they can fabricate it to place it at the scene of a crime, thus implicating you. http://www.nytimes.com/2009…more broadly, though, many would probably agree that we are headed towards a world in which biometrics will increasingly be used to authenticate. i would suspect DNA information could thus be a vehicle for identity theft as this trend progresses.
biometrics isn’t the same as DNA, and what would identical twins do?
Charge up each others credit
There is one situation where it could be a concern, middle and upper middle income folks who use life insurance to protect their families from lost income.There are no federal or state laws that prevent a life insurance company from discriminating based on your genetic data.
Maybe the reason you don’t care is you have no hereditary disease in your family history (I don’t know you, so I have no idea) and you’ll never have to apply for a job (I suspect that’s true). I’d be concerned that a self-insured employer or one with an insurance plan that is capitated in one form or another might discriminate against me if my DNA contained disease markers.
Some people might not want all the schizophrenia genes in their DNA on public display for employers, insurance companies, potential romantic partners…just sayin’
I think the divide here is between judgmental and non judgmental people. If you aren’t judgmental my guess is that you tend to think that everyone else isn’t either. Hence you ignore a potential danger because seat of the pants it doesn’t appear threatening to you.You know back in the day companies used to interview the wives to make sure that they were subservient enough so that the husband could get his job done (with full support) unimpeded. And the reality of hiring someone (for lack of a better way to put it) “with problems” is something that is very very real in a traditional small business where the full load falls on very few people. If one of those people isn’t able to function it falls on “mr owner” as problems and aggravations. It makes 100% sense that you would want to avoid problems down the road as opposed to just being a nice guy and clicking your heals and wishing that everything will be fine. Perhaps this is something that those who work for corporations and have job safety nets don’t fully understand.Which is why having government policies to encourage or force companies to give family leave (as only one example) are a mistake. Even if it is something that only large or financially secure companies end up having to do, it still greatly disadvantages the majority of small companies in insecure positions that aren’t in a position to offer the same benefits. Hence they will never grow to be larger companies.
http://www.nytimes.com/2016…”People who carry genes that accelerate or intensify that pruning are at higher risk of developing schizophrenia than those who do not, the new study suggests.”Current understanding points to genes not being sufficient…environment triggers the gene expression…if one identical twin develops schizophrenia, only ~30-50% likelihood of other twin developing it……babies born after a bad flu season or e.g. Finnish Winter War trauma more likely to develop schizophrenia…so it seems quite likely that people who didn’t suffer from the disease would get the stigma of carrying the gene if information were widely available like a credit report.accusing someone of being judgmental in this context is frankly insensitive. it’s like telling people who have been victims of e.g. racist police brutality that they see racism everywhere due to some character failing in themselves.to me it seems naive to the point of ignorance to ignore the inevitability of this information being used in socially undesirable ways. (unless one sees eugenics as socially desirable) what are you going to tell the kid who got barred from school because of a gene? is there something he doesn’t understand because of big guvmint? http://www.wired.com/2016/0…
“As genetic testing explodes, health insurers are banned from denying coverage based on results. The same doesn’t apply for life insurance.”http://www.fastcompany.com/…
Fred, if you are open to publishing your data, I would also encourage you to check out https://www.openhumans.org/, which will let you sync and publicly publish your 23andMe data, so that scientists can have an open data set to work from. The site is an offshoot of the Personal Genome Project, a Harvard project that is paying to fully sequence genomes (much more genomic data than 23andMe provides, but currently $1k a pop), so long as you publish your data openly.(If you are familiar with the BRCA (breast cancer mutations) court case, the Supreme Court held that certain patents on DNA were unenforceable, now allowing other providers other than the company that held the patents to test patients for those mutations. But that company still has 10+ years worth of genomic data and health results, and thus, even with the patents invalidated, may still have a big advantage in how to interpret that data diagnostically. An open network of public genomes and health surveys can move science forward outside of corporate silos.)Also, unless they removed the feature, you should be able to link yourselves as family on 23andMe, and your daughter could see what DNA came from which parent.Another interesting tool is promethease.com, which lets you upload your 23andMe raw data and cross-reference it with SNPedia (a wiki for interpreting genomic data). Since the whole FDA fiasco, 23andMe can no longer show most health results to new users, but promethease still can since it is not providing the underlying interpretations, just cross-referencing SNPedia with your DNA.
Seems like consumer genotype data is fairly low-resolution at this point, nowhere close to a full sequence. I don’t think you have to worry about someone planting your genetic prints at a crime scene at this point. Maybe if you have a heritable illness you don’t want to make public, or an unacknowledged child…
full sequencing costs about 220k/individual.
23andme is founded by two women, Anne Wojcicki and Linda Avey. It was a woman, your daughter Jessica who encouraged you to get the family’s genes sequenced.Today happens to be …
I thought about this long and hard when I submitted my data to 23andMe about 4 years ago and I came to the conclusions that a.) This data is valuable b.) I can’t fully trust 23andMe. I then thought about c.) All this can be done with a swab of my cheek or a stray hair I leave almost anywhere by someone nefarious. I felt (and still feel) the value of having it done outweighed a & b.
Good points. Although I think there is a notable difference between “anyone can grab some of my hair” and willingly submitting yourself to a genetic database.Everyone has their own comfort zone for sure.
I think that privacy is big depending. I have a friend who told me to always pay cash at the doctor. Made sense for me because I had a $10k deductible anyway. That way insurance companies have zero access to your data-which is the real concern. If you knew my genetic make up, little you could do with it except send me ethnic food from Norway, England and Germany.
siding with jess in this beef. i’m very interested in 23andme for my family, but the error rate, coupled with the privacy issues which i regard as very significant, leave me unwilling to try.
in olden days, when pre-existing condition meant no health care for you, it was potentially a much bigger issue than now that insurance / health care isn’t ruled out; there are still some dystopian sci-fi scenarios about genetic discrimination in the future.
How many times have the republicans voted to repeal Obamacare? It’s also a campaign rally for the current front runner. So, I wouldn’t take that issue off the table just yet.
In May 2008 the Genetic Information Nondiscrimination Act (GINA) was passed 95-0 in the Senate, 414-1 in the House, and signed into law by George W. Bush. No opposition, bipartisan, and morally right. To some, it seemed inevitable that a pre-existing conditions clause would be next.Please don’t confuse folks’ distaste for Obamacare as a desire to strip the “pre-existing conditions” clause from our list of civil rights. We have direct evidence that when laws like this are brought to the table unhindered, they can pass with ease.When those rights are cynically used to buoy a complicated and unpopular 2,000 page social experiment, they simply become powerful ammunition for attacking the law’s opponents.
Can you imagine if someone that was viewing Fred’s info & found some abnormality and could actually alert him to go get a professional diagnosis. I believe there is an HGTV show where some viewer noticed a strange lump on the host of the tv show and sent in a letter that urged the host to go get a biopsy. This viewer potentially saved someones life. The power of an audience.
The genealogy and linking stuff is there already (at least on the UK version), you just don’t get to use it until your profile has been done.
Agreed, as soon as your profile is loaded it will automatically update your daughter that she now has two direct genetic relatives on the site, and obviously you and your wife will both have one each in the former of your daughter ☺I have the old USA version prior to the FDA lock down, although it was ordered from the UK where I still live, and this gives a much broader range of results than the more recent versions, I can also have visibility of family tree making websites that they are in partnership with, which I would have assumed would be available on the American version too?You can also download everything that was sequenced by 23andme and load this data in to other sites such as Promethease for even greater health detail than is displayed in 23andme with links to relevant published papers based on your genetic makeup.
Fantastic. Can’t wait to see that
I too like the idea to share when everything is OK, but…Have you thought of the Negative Impact if something in the data shows that not all is OK ?Let’s say that user ALPHA discovers that his brothers are not of the same Family ?Or that user BETA has a genetic late life serious disorder that has been transmitted to all his children with not cure available ?Or… or… 😉
Would it really be possible to discover that you and your siblings had different fathers? What if the fathers had similar genetic backgrounds? (Asking for a friend 😉 )
hmmm, what if you do not have the same father? #couldhappen 😉
Oh, I’m counting on it. ;-P (Word to my brother!)
Should be a choice if one wants to participate. That said, this could be critical for preventative healthcare and reducing treatment costs. If someone realizes they are predisposed to x condition – they can take the necessary precautions. But yes – realizing you have a 95% chance of getting cancer could reduce your enjoyment in the present and lead to dread of the future.
and what if you could take no precautions ?
I’m more concerned about Jess and the GothamGal.By default, since the GothamGal is ashkenazi Jewish, she’s the product of a lot of inbreeding from intramarriage for generations and a number of founder events (and frankly, so am I). The overall carrier risk for anyone fully ashkenazi for a recessive is about 1:2 (though which disease and what carrier risk varies http://www.jewishgeneticdis… ) For Jess, that would be ~ 1:4 (fred is not ashkenazi, though he can also carry recessives).There are new recessives and other nonsense being discovered all the time (https://en.wikipedia.org/wi… I know people who know the parents, aI grew up similarly to the parents, they would have genetically paneled against each other before getting married, and they came up clean. Their child who died is a genetic variant discovery)All of Fred’s kids are young and have their lives ahead of them. Having recessive carrier status is not ideal at all.
I was making just a general statement… #DoWeLiveBetterIfWeKnowAll?
I am the most wrong person to ask here. Some days I feel like I could throw up over that question. (because I don’t want to type up the full story again, go to the two links)http://avc.com/2016/02/23-m…http://avc.com/2016/02/23-m…I don’t know I don’t know I don’t know, and if I knew I would know what to tell you
Just did 23andMe a few months ago. Very happy to share all of my genetic data and hopefully have it contribute to science. I can’t wait until I can share this information with my physicians.One area that they did poorly was identifying my background – according to my genetic makeup, I’m only 8% French. When I shared this information with my grandparents, they pointed me to this massive family tree that traces our roots in France back to the year 1100! After looking more deeply into 23andMe, it turns out that their screen is highly inaccurate for French and German genetic markers. This information was well hidden and I wish there had been more transparency.
Believe the results and data get more accurate as more people participate.
I signed up for 23andMe several years ago on impulse during a discount promotion they offered. I bought my mom one too since she’s been researching our family ancestry for years. There was nothing shocking in the health results. Just a few unconcerning things. What was really interesting to me was what they can do in terms of tracing your lineage back to certain regions of the world and strong blood lines. I hardly ever refer back to the data, but thought it was totally worth the price if only for a one-time set of insights. My mom even found a few distant relatives through the service.
Should 23andMe do this, or some third party with the sole purpose of connecting the dots? Kind of hoped HealthKit would have solved some of this. Still quite an opportunity.Side note: some of the things I enjoyed most about my results were the trivial things like the unlikelihood of me being a good athlete (hope to outsmart my genetics on that one) or distaste for Cilantro. Reminds me that, in most technologies, novelty precedes utility.
Mrs. Dr. Jules and I did 23 and me for Cheistmas, and found it fascinating. I totally agree that it is missing hooks into family tree software…. Though there is a clunky sharing feature.I am related to Congressman Nolan from Minnesota – a really nice guy. Turns out we are both related to Lt. Governor Bill Nolan from ~1910. A newspaper report from the time… “Lt. Governor Nolan, a known drunkered and brawler…..”Congressman Nolan reports having travelled to Ireland, finding that Nolan’s are one of the oldest clans in Ireland, though not much is written about them, having failed to doing much of anything noteworthy in the past 1,000 years. Oh well, at least I married well.
Great post, Mike
I think there is a way to link them, but I haven’t done it yet. Go to this link, and it’s under the heading, “Close Family”. There’s a mini tree.https://you.23andme.com/too…
I agree that a lot of the power/fun of doing this testing is in being able to share it. My company provides dentists a protocol for personalized education, treatment and product recommendations (with home delivery) and we’ve been thinking about adding a genetic testing component through a partner. I think there is an opportunity to get more people involved in their healthcare by making it easier for providers to recommend the testing. Building a stronger doctor/patient partnership being the goal.
So, you’re advocating that family and other data should be connected more naturally and organically. That makes a lot of sense because it gives better insights.Here’s what I’m interested in learning. What’s the “So, What” of this? Better outcomes is an old cliche that has been used for a long time.- Are there specific decisions or changes that this data has allowed you or your doctor to make? (without disclosing confidential stuff of course).
Great questions.- Are there specific decisions or changes that this data has allowed you or your doctor to make? (without disclosing confidential stuff of course).As mentioned in my other comment the other thing that is important to factor in is the anxiety of knowing that something might happen in the future and in particular when there is nothing realistic that can be done to prevent it.
i expect FB will enter this space in the very near future.
You’ll be able to connect as a family once your results are in. Also, take a look at this article regarding 3rd party services that utilize the 23andme API and provide greater insight into your results (with the caveat that they may be somewhat misleading): http://www.slate.com/articl…
Guessing 23 and me might not want to get involved in “certain situations”….
I came here to say this. I recently read “The Juggler’s Children” – a book about doing genealogical research with a strong focus on genetics, and apparently the “false paternity” rate is something like 10%. That’s a pretty big mess to get into.
If you think about it, we are all walking data factories. From a business perspective, I believe the next frontier in data is harnessing, using and monetizing all of the data we create as individuals. Think of it as the convergence of “big data” and “internet of things” (terms I hate by the way) — I want a personal API of all of the demographic/medical/recreational/commercial data I create, securely stored in the cloud, that I own and control.
I think a consumer data Union where you could fully control your data and even allow for the ability to share for financial benefit (nice basis for basic income) would create the platform for far more data, more sharing and far more advances in health care and other verticals
I have a friend who is a genetic counselor, she recommended me to use Promethease which does a more comprehensive screen via the 23andMe API against tons of genetic research and studies.She also warned me about the issue with third parties providing services and additional data… it’s a gray area providing average people with such deep knowledge of their health. Inevitably, you’ll always find bad news the harder you look. It’ll be an interesting ethical debate moving forward how much knowledge 23andMe will disclose to its customers.
Check out TapGenes
Rumor: 23andMe being bought by Ancestry.com
The investors won’t make out on that one…
Healthcare is a classic case where government regulations essentially limit sharing as much as technology does. I’m pretty sure that as long as you (the user) has authorized the sharing, it would meet with HIPAA security and privacy rules. But when you start talking APIs – Oh, my!Similar to other comments, I have always wanted to write a universal “sharing” app that keeps a detailed record of every site (organization) I have authorized to store or process my personal information and exactly what information. This would address both data privacy and transparency issues in one platform. (I think that is what Google health was supposed to be?)Oh to be a twenty-something hacker again! I’d have this done by Monday.
I have a very close friend (wrote a book: http://www.amazon.com/The-C…. HIs son recently was diagnosed with a brain tumor. They did genetic testing on him, his deceased father, and his brother to find out if they all carried the same cancer gene. One son did (the one with the tumor) and I don’t know the results of the other son. Women that have a history of breast cancer or other cancers have done this too. Sometimes it’s worth it to be proactive. Sometimes, there is nothing you can do and there is a time bomb sitting in your body somewhere. But, it’s not only cancer. Would be interesting to see if you were predisposed to things like diabetes. Much of health risk can be minimized via lifestyle. It’s a totally interesting space and AI will have a lot to do with it in the future.
Sometimes, there is nothing you can do and there is a time bomb sitting in your body somewhere. But, it’s not only cancer. Would be interesting to see if you were predisposed to things like diabetes.I think the anxiety factor as a result of knowing certain things in advance is important to factor in to a decision to do testing like this.
I don’t want to think about this too deeplyWe don’t understand the mechanics of cancer and genes well enough at all to minimize the risks enough
Ancestry is actually working on health stuff http://health.ancestry.com/ . The finding relatives has proven really helpful for me as my dad and his brothers disconnected from family, so it helped a lot with finding out who came from Ireland. FamilyTreeDNA does a much deeper genetic geneology trace that allowed my clan in Ireland to prove me a direct decendant.
Responding to the few threads about data breaches…I ran a healthcare business that digitally stored lots of medical information on lots of people, primarily results of STD tests.While there is always the risk of inadvertent / self-inflicted breaches, the practical risk of a deliberate, sophisticated attack is small. There is no economic incentive to compel cyber criminals to act on you. Theoretically there is “ransom” potential, but that’s way too little juice for the squeeze compared to the economic incentive associated with attacking an e-commerce site and all their credit cards.There will always be rogue actors that want to make a political statement, but if a health care site uses just the industry-standard protections used by e-commerce sites they’re likely fine.I used this argument successfully when raising three rounds of venture money. My sound bite was there is likely a 1000 to 1 difference in the volume and sophistication of attacks for an e-commerce site relative to a health care site. And we employed IT staff that had protected e-commerce sites, so we were like fighter jet pilots operating a Cessna.
There is no economic incentive to compel cyber criminals to act on you.People often do things just for fun with no economic reward. Hackers in particular fit this mischief role. (Think of graffiti in the city as another example w/o economic reward).
Totally agree. I probably buried my lead on this point. There is certainly a risk…but the point i was trying to make was the industry-standard protections used by ecommerce sites, when applied by healthcare sites, should be more than enough to stop the mischief-maker attacks. The “no economic incentive” argument applies only to the sophisticated, coordinated attacks.
I’m not sure this argument would still fly. In fact, the data shows that healthcare data is 5x more valuable than credit cards in a breach. I agree though, about using best practices. The problem is that 90% don’t!
Would be really interesting if they could some how anonymize all of this data and make it public for research, etc. Probably some big time ethics issues but there could be a significant amount of “good” that could come out of this in terms of advancing the understanding of the human genome, creating more cures…. but also bad. There may be other global databases of data like this, just not super familiar.
They already are. 23andMe has contracts with several Pharma companies.
Curious about the anonymity of 23andMe though. For instance, in the case of closed adoptions, parents and agencies maintain confidentiality and sometimes don’t even disclose their identity. Finding genetic makeup on yourself or an immidate relative is one thing, butfor the same closed adoption reasons- knowing the risk of allergies, genetic mutations or diseases or other things that “run in the family” would be helpful for parents who’s adopted child gets sick or other. I agree with most here that privacy is an issue.Perhaps connecting to others could be solely opt in?Side note: my dad has been creating our family heritage tree for nearly a decade now and it’s intense and really fascinating.
information wants to be interoperable
Nice. Updating Stewart Brand :-).
We stored cord blood at child’s birth as step one in this process. I wonder how many people do that.
skipped it … i doubted it would be there 20 years from now
genecoin anyone? publishing your dna to a blockchain is powerful.not everyone might want their family gene code profiles mapping and linking. it could create awkward moments.
Maybe my 23andme is different because I had it done before the FDA clamped down on it but there is a ‘Family & Friends’ tab at the top where you can add DNA Relatives, build a ‘Family Tree’ (affiliated with MyHeritage.com though not Ancestry.com) and see DNA traits that are passed down thru generations. My parents have not done 23andme but my sisters have and I was able to add them as siblings and share DNA with them. When you agree to share your genetic information with someone – the other person has to consent as well – you see their genetic results along side yours for every trait (hair colour, cilantro gene….what have you).Under my Family & Friends tab I have both my sisters at the top and then a bunch of other people I don’t know that 23andme has classified as 3rd to 5th or ‘distant’ cousins.
Fred – The option to connect to family members does exist – it happens after your DNA is sequenced. That’s the magic part: imagine, if they can automatically tell you that your daughter is your daughter without you telling them that to begin with, that’s when science leads to awe. The system will also start suggesting relatives up to the 5th degree of separation and you will be able to build family trees and share your genetic data with everyone. You are correct that what they are still missing is the ability to really sinc all this data up with your healthcare provider, but they are already working on it. The utility of the results will indeed increase over the next couple years, especially as medicine and drugs become personalized instead of mass-produced…
23 and Me will build that relationship for you automatically. I did a kit a long time ago. Then, a few years later, when my mother did one, it alerted me that my mother had joined without me having to do anything. So I’m guessing both you and GG will be connected in that you are the genetic parents of that person (it won’t and shouldn’t probably care if you’re married or not).Another interesting thing happened with 23 and me. It’s a long story, so I’ll spare the details, but there was a family legend that was completely disproved by my genetic results. In essence, the lineage some members of our family were claiming is not possible– at least not if we are actually related to each other 😉
There’s one regulatory issue that people should consider before signing up for this service. The Genetic Information Nondiscrimination Act of 2008 (GINA) prevents a healthcare insurer from imposing a preexisting condition exclusion based solely on genetic information. So having a genetic predisposition should raise a rate or become excluded from coverage. However, this protection is not extended to life insurance policies. So there is a very real risk of people taking this test and then getting denied certain levels of life insurance due to the risks suggested by the test results. More from 23andMe here: https://customercare.23andm…I’m also concerned that the GINA protections haven’t been tested in court or with the regulators as of yet. So there is some risk of taking this test and finding out at a later time that it effects your health and life insurance coverage.
Thanks for the great post! Be sure to check out the ‘Tools’ section of the site to share & compare genetics with your family once your data comes back: https://you.23andme.com/tools/At 23andMe, we deeply believe that individuals have a right to their own data and can move it at will. To that end, we’ve offered ‘raw data download’ since our launch in 2007. We’re also huge believers in the promise of better health data interoperability. We publicly launched our API several years ago and have been pleased with the level of interest both from our customers & from the development community. Consumer-facing sites like MyHeritage, physician-facing (like http://projects.iq.harvard…. or MDRevolution), and research-oriented apps are all successful consumers.However, as an FDA-regulated device, the regulatory path for building & promoting our own appstore in the US is a bit challenging at the moment. Feel free to reach out if you or your portfolio are interested in collaborating.
Fred, really appreciate the feedback, I passed it on to our customer care team. There is a way to connect with family on the platform, more on that here: https://customercare.23andm…. If you have trouble you can email me at [email protected] and/or I can put you in touch with our customer care team.
The larger point — what I believe Fred is saying — is that it should be the individual’s choice. If he wants to publish his data and link it to other apps, etc., that’s his choice. Others needn’t do so. It should be their choice. Handling the data securely is a crucial but separate matter.
Fred- interesting post. It’s great to have active do-ers like yourself making suggestions about improvements in health care/tech companies. That space has suffered from so many issues in terms of helping patients, doctors and hospitals communicate better with each other that sometimes I despair that it will ever get fixed. When you think about that, this is tragic because there are millions of lives lost every year because of information/communication problems in this area.
Neanderthal and me;https://www.theguardian.com…
Fred,If you go to https://carebox.it you can set up a Carebox account which supports the 23andMe API and its OAuth2 authentication for connecting your 23andMe account to your Carebox account. You can then combine into one cloud-based location your 23andMe data with that from any of the hundreds of patient portals that can be connected to your Carebox account (including PatientFusion portals), medicare data, VA data, imaging studies you can upload, and more.
Could not agree more with the first piece of feedback. Just gave the exact same piece of feedback to a friend running a personal finance app, also a team sport. I think the service provider’s privacy fears/concerns are so high they just steer clear altogether. Maybe not a law, but a norm.
I believe 23andMe intentionally doesn’t do some of what you are looking for right now…in fact, if I remember correctly, they actually used to give you a whole lot more information when you got your results, but because of ‘legal issues’ they now give you less (and let you do less with it too).I’m sure they are anxiously awaiting the day when they can (legally) start to do much of what you asking for (and go back to giving you even more complete health results from the genetic testing)…the challenge for them really seems to be in “what to do in the middle” while they wait and push for a more open (and less legally inclined) world…
I’d like a copy of your genetic makeup. Perhaps there is a patent I can get for some medicine based on it or if you have a good characteristic I could put that into a child of mine. Thx! Personally my belief is that the power of technology will allow individuals to create destruction on a grand scale; and I’m good at algorithms and SW, so gathering the genetic data sounds like a productive big data acquisition for me.
Did 23andme a few months ago. Information is interesting, but their web app was kinda crappy, the UI improved a bit in the meanwhile but it’s still _very_ confusing…They seem to promote MyHeritage instead of Ancestry.com.There’s this option of adding other 23andme users as your relative https://you.23andme.com/too…
I got a kit 2 years ago, had to mail it back from CT b/c it was (maybe still is) an “unlicensed lab test” in NY. They could ship it to me here, but I’m not allowed to “administer” (ie, spit in a tube) within this state apparently. Results were a lot of fun for a few months, best value for $99 I ever got.
Most likely hamstrung by meddling middle men known as bureaucrats.
I’m adopted, and know next to nothing about my biological family. The insights regarding things I may be genetically predisposed to that I otherwise would have no visibility towards was well worth the price.
Wow, yeah. Not to trivialize it, but it sounds like a great basis for a novel or movie.
It dawns on me now that, at my age, I’d like to do this to see if I’m carrying any health-related genes that I might be able to address before it’s too late.
Full Disclosure:I really can’t talk about parts of this subject neutrally at all. I’m facing if/when/how(including if I need to recruit more family members)/what are the repercussions long term/ for doing certain kinds of genetic testing, namely around breast cancer (and possibly many other cancers as well). I’m one of the few individuals who are extremely high risk, with a mother who had breast cancer and does not carry any sort of BRCA mutation at all, and an aunt (her sister) who also had breast cancer and is a carrier of a known (ashkenazi) BRCA1 variant. We know where my aunt’s variant came from (my grandfather, who is still alive, has that variant). Their mother was diagnosed with breast cancer around age 32, and died of either breast cancer that metastasized to the bone, interrelated bone cancer, or unrelated bone cancer when my mother was 18. It is likely that I am a candidate for whole genome testing as an archive/reference individual to start shooting in the dark for more links about how cancer works, and that’s BEFORE premarial Ashkenazi genetic testing for carrier status of a huge variety of disease. (which is also on my plate now). It is extraordinarily rare to find individuals who are high risk, definitely ashkenazi, and almost definitely not BRCA carriers. It is highly probable if I enter a study, nothing major will be found in the short term. In the long term, am I looking at the possibility that I am looking at some new form of later onset Li-Fraumeni Syndrome? What am I supposed to do then? 23 and me is only useful to me insofar that you can download your genetic information that they have on hand. I find it really strange that the medical community and research community in some cases can have more access to my genes than I can if I became a research subject, and I kind of want a copy for myself to ponder.For further thought: Apparently there is a market within the research community for human genome sections, especially if it has family histories and medical histories behind it. I’ve heard estimates that pieces of my genome on the low end is worth between 1-5k (but not to me, and I can’t sell it.)(and supposedly my mothers is worth a lot more, especially in conjunction with my aunt.) Again, slightly freaked out by the thought. That’s its own set of ethical questions, especially since 23 & me does participate in broader research.
And I wish I could just test and authorize to research via API as two independent actions from each other. I’d be a lot happier
What would torment me the most would be not knowing where I stand. Perhaps you are right, ‘perhaps’ there is a market for any of our data, realistically we can’t control that and it doesn’t bother me, there must be some positive outcome for the benefit of all, some progress based on our data (alongside revenues for various companies). The value you would extract from this test, for you personally, is still huge. I have watched some testimonials on youtube and will take the test with first opportunity and if I was a mother I would see that as a responsibility towards my (future) children.
So I’ve talked to researchers in the areaThere is a market. That I can’t participate with the researchers as an equal partner does bother me ethically. I do not want to be a Henrietta Lacks( https://en.wikipedia.org/wi… ), except for genetic pieces.As someone where we are near positive I’m not a BRCA carrier at all despite being very clearly and traceably askenazi, (we did in depth family trees, BRCA is dominant, and my father’s immediate line doesn’t seem to show a trace) the value for me starts getting into testing for other secondary genes like PALB2, ATM, CDH1 (unlikely for various reasons, but would be checked for anyway), RECQL (unlikely), RAD51, CASP8, BRIP1,EMBP1, CYP1BI-AS1, MDM4, CASP8, SLC4A/NEKID, MAP3K1, FGF10/MRPS30, TERT, ESR1, LRRC4C, ZNF365, FGFR2, LSP1, PTHLH, TOX3/TNRC9, BRCA1-wild type, BABAM1, GMEB2, FGF13, TGFB1, (CHEK2, p53, -> Li–Fraumeni syndrome, unlikely, but if I have it, i’m probably dying by 60 something), (EPCAM, MLH1, MSH2, MSH6, PMS2 -> lynch syndrome, unlikely, but there)*.They probably would also throw in BRCA1 and 2 just to be sure, but, as I said, almost every doctor, genetic counselor, and researcher I’ve spoken to is pretty sure I don’t have carrier risk (father’s immediate line doesn’t show a clear reason why they would be carriers, mother’s line would except that my mother is explicitly known to not be a carrier, both are ashekanzi).The problem with this kind of testing is:a) we don’t get enough samples of people in general (statistics), so what’s known as a VUS (variant of unknown significance, aka this is either the first time this is seen, or they haven’t seen this variation often enough to say anything about it) is very normal for even straight BRCA 1 or 2, and tells you nothing without context, and especially tells you nothing for support genes by itself without large samples.b) most are supports for BRCA1/2, which we really don’t understand the mechanism of how it works in terms of cancer causing/or not. (like its presence makes you at risk, but it may not be the cause, the cause could be something totally different, which ends up triggering the fault gene, which is a totally different issue and which means this process is f***ed up, and what is the process means a totally DIFFERENT GENE! that we don’t know about is triggered which triggers brca genes, and that gene is triggered by epigenetic causes. Then what? See the issue?)c) Remember how I said earlier there is an assumption about how I have little carrier risk for BRCA1 or 2 – well if I did come up positive, then we would guess it came from my father, and most likely it would be a VUS. Given that nearly all the risk is coming from my mother’s line, interpreting that VUS would become a disaster because of b.as a result, I basically expect either nothing or lots of VUSes, which basically also mean nothing in the short term, because they could be a sideshow to the real thing to look for as far as the world knows. I fear coming up positive for Li–Fraumeni syndrome if anything, but that is really really rare.I’ve also been explicitly told to not expect much information because of my background from these tests. It is rare and super unfortunate when that happens on one level. OTOH, it also usually means you’re flagged as a gift to science. Which also makes me feel a bit like a hamster.Now where is my wheel.* http://www.cancer.gov/types… and http://www.cancer.gov/types…
hi Shana, I am glad to read your reply because I mistakenly understood that you prefer not to know what may be your future health. I have opened the wikipedia link and it turns out that HeLa cells ‘represented an enormous boon to medical and biological research’. I think none of us can be an equal partner with research/science, realistically it has always been like that and we all benefit today from thousands of years of trials and errors and ultimately, progress. In conclusion I think it is good we have today a choice to know, it is early days and not perfect, still another step ahead compared to previous generations. I wish everyone took such an active interest in their health as you do.
They are, but at great ethical cost to Henrietta Lacks and her family.
Tapgenes does what you’re asking for.
Two 3rd party apps I found quite good. Athletigen (amazing), and promethease. The second one is very detailed medical info and costs about $5, was more useful when 23 and me didn’t give you health information. The former looks for genes known to be correlated with many aspects of athletic performance (including response to training and even willingness to train). I actually think if I had taken it when I was 14 it would’ve given me all kinds of new things to try, but could also see it being discouraging if some of my results were worse. Probably get over it by watching Gataca.
Not sure if someone else mentioned this, but I don’t think 23andme actually sequence your DNA. I’m pretty sure they use SNP chips to identify the presence of specific genes.
Check out https://www.promethease.com. You can pipe your 23andme data there to get some expanded reports on your SNPs
Indeed, I read a few years ago that a high percentage of people have different biological parents than they think they do. This was in an article in the NYT regarding paternal custody rights. It turned out, in this particular article, that a high percentage of Dads turned out not to be the biological fathers.
Hence the trying to burying the facts under information overload approach hehe.Fortunately, I remain 99.8% Northwestern European. My grandparents did not stray too far
Yeah but doesn’t that require samples from at least one paternal candidate? I was curious if DNA testing, alone, on siblings could reveal different fathers. Seems unlikely to me unless there’s a really obvious marker in there.
That’s interesting. My first reaction was to think of a pejorative around the mother. But then I realized not only was that not fair but also perhaps the couple on purpose used another man’s sperm because the father was sterile. And then they did what was practical which is not tell the child.In judaism for a child to be born jewish the mother has to be jewish the theory being that you always know who the mother is.