Consumer Genetic Tests May Have a Lot of False Positives (theverge.com) 99
A new study, published in the journal Genetics in Medicine, found that consumer genetic tests bring up a lot of false positives. "In this case, 40 percent of the results from the consumer tests were false positives," reports The Verge, noting that the findings "cover a very small sample size and don't show that consumer tests always have a 40 percent false positive rate." From the report: The research was done by scientists at Ambry Genetics, a medical laboratory in California. By looking through their own database, they found that 49 people had been referred to them because of some worrying results from their consumer genetic tests. Still, scientists at Ambry were able to confirm only 60 percent of the results when they compared the raw data from consumer tests with more thorough genetic tests done by themselves and other clinical laboratories. So, 40 percent of variants in a variety of genes reported in DTC raw data were false positives, meaning that they said a genetic variant was there when it wasn't. (Most of these turned out to be variants linked to cancer.) Additionally, the authors write, some variants classified as "increased risk" were not only classified as "benign" by clinical laboratories, but they were actually common variants.
Re: Is a back door for law enforcement (Score:2)
They don't have to. The multiple layers of anonymizing means nobody can provide anything useful. Besides which, there's nothing in there the government can use.
Paranoia is a psychiatric disorders, not a political stance.
Re: (Score:2)
Do they accept anonymous chunks of gold?
I once bought something even though I wasn't the end user. I once bought something with a prepaid card I paid cash for. I once received a gift.
Re: (Score:2)
Re: (Score:2)
At no time did you have to provide your personal data.. You certainly can. But it's not required.
No worries - six cousins have already submitted theirs.
And you bought the access ID with your MasterCard.
Re: (Score:2)
Doesn't matter, their samples aren't attached to any personally identifying information, and I can buy a testing kit for anyone. I've bought testing kits for several people. Doesn't mean I'm any of them.
Re: (Score:1)
Do you sincerely believe these consumer tests have been so thoroughly anonymized that it can't be undone? Or that they're actually doing it at all? I sure don't.
Oh, really? "We've recovered DNA evidence from a crime scene, we need to search your data for matches and relatives". If you don't think governments can and will abuse this, you're missing
Re: Is a back door for law enforcement (Score:2)
DNA can only be used to exonerate, not convict. Law enforcement might use it to identify potential suspects for further scrutiny, but when it comes to a court & jury, the only thing DNA proves is that it either a) could not possibly have been you, or b) might have been someone related to you.
Re: (Score:2)
The DNA evidence they get constitutes 12 markers that aren't included in ANY of the DNA tests provided off the shelf. It's also so thoroughly discredited, because too few markers are used, that it has virtually no legal value any more.
Re: (Score:2)
Do you **REALLY** think these companies are going to put up a legal fight for you when your DNA is requested by the government?
A.C.#2
They don't have to. The multiple layers of anonymizing means nobody can provide anything useful. Besides which, there's nothing in there the government can use. Paranoia is a psychiatric disorders, not a political stance.
If you cared to Google it, you would find that, with a search warrant, law enforcement can obtain genetic information and material from places like 23andMe and Ancestry. In fact, the link below confirms they already have at least once. You can use a pseudonym. Maybe it and the corporate anonymity maze will be enough to slip through the sophisticated clutches of the Law's big data algorithms. GLWT.
https://www.ajc.com/news/national/can-police-legally-obtain-your-dna-from-23andme-ancestry/8eZ2 [ajc.com]
Re: (Score:2)
They get a search warrant. Their system shows that there's a few million test tubes storing DNA samples. That's more than the budget for all the police departments in the US.
They obtain the SNP values. Useless, the police database doesn't store SNPs, it stores STRs in one of the less interesting chromosomes. No way to compare the data.
They use an archaeology DNA lab to sequence the crime scene sample because improper storage means it has broken down. It comes up with about a thousand results, because DNA la
Re: (Score:2)
They get a search warrant. Their system shows that there's a few million test tubes storing DNA samples. That's more than the budget for all the police departments in the US.
The issue addressed wasn't how expensive it is to properly analyze DNA, but whether companies like 23andMe and Ancestry will turn over your DNA to the government. Yes, they will and have.
They obtain the SNP values... They use an archaeology DNA lab to sequence the crime scene sample because improper storage means it has broken down.... DNA lasts upwards of a million years and there's a lot of cross-contamination... They're not methodical, they use DNA to try and rig conviction rates...
Archaeology, archaeogenetics, what do I know? I'm not going to put forensics on trial. PBS's Nova already did that: http://www.pbs.org/wgbh/nova/tech/forensics-on-trial.html [pbs.org]
Your average American would prefer fake trials to paying for a decent service.
'Criminal Justice' is a double entendre. For some reason, this brings to mind the Nietzschean allegory about a murder. The sheriff discovered
Re: (Score:2)
Yes, I know that's satire, but there are a few things that people might worry about.
Only partially for hair colour. I'm something like 75% brown-haired, 20% red-haired, 5% black-haired. What's law enforcement going to do with that? "Someone who may or may not have hair, which may or may not be absolutely anything at all" is not much of a fit. Epigenetics and diet can alter the ratios. Similar problems exist for eye colour.
Race is falsified by DNA. You carry mutations from just about every human (and several
Re: (Score:2)
Then you've (a) never used a DNA service, (b) never bothered to look up what the police DNA database uses (clue - it's incompatible), (c) don't understand DNA.
You cannot de-anonymize the results, don't give me that crap.
The fact that you can find a criminal pretending to offer a service (and apparently those are the circles you work in) does not make any difference to the status of legitimate services.
DNA cannot be abused.
I probably know more about the last two decades than you do. I certainly know more abo
Re: (Score:2)
If individual-level genetic data is available (as is the case for at least 23andme), then it can be de-anonymised.
Dr. Erlich also identified a new genetic privacy loophole that allows inferring surname of individuals from simple Internet searches using genetic data.
http://datascience.columbia.ed... [columbia.edu]
If you have individual level genetic data, fewer than 50 common variants should be enough to uniquely identify a person.
Re: (Score:2)
It isn't. Ignoring the fact that you're a chimera and a mosaic, which means you can have multiple combinations of those, we know from genetic genealogy that 111 markers will be sufficient to uniquely identify the group that comprises every relative up to three steps away (so third cousins, great grandparents, etc).
You'd need far, far more markers to uniquely identify you. And that's useless as your DNA will last up to a million years. Cross-contamination means that unless you can identify how old the DNA is
Re: (Score:2)
It isn't. Ignoring the fact that you're a chimera and a mosaic, which means you can have multiple combinations of those, we know from genetic genealogy that 111 markers will be sufficient to uniquely identify the group that comprises every relative up to three steps away (so third cousins, great grandparents, etc).
Fine, if you don't like my 50 common variants number, then I'll suggest 120 variants: 111 [oddly-specific] to get down to familial group, and another 10 or so to identify a single person within that group. Whether it's 50 or 500, that's still well within the realm of cheap targeted SNPchip technology.
You'd need far, far more markers to uniquely identify you.... You'd need full genome sequences from multiple collection spots across the body, plus sequencing of the sample, for that.
There's a big difference between uniquely identifying someone, and fully describing their genome. I agree that a full description of a person's genome would require extensive whole-genome sequencing, but that's
Re: (Score:1)
Re: (Score:2)
Do you **REALLY** think these companies are going to put up a legal fight for you when your DNA is requested by the government?
Why not send your sample to an overseas lab? Surely somebody here has one to recommend.
Commerce in health (Score:5, Interesting)
There's a massive incentive to *find something* if you're paying to have a check done... if you get nothing reported then the value of the test seems to be nothing. If you are told to be "at risk" for some condition then you can go to your friends and tell them that you had no idea and that it was so worth it and that now you can take steps... and that perhaps they should get tested too. The best part is that if you're just reported to be "at risk", rather than an actual diagnosis, there's almost no need for accuracy. If you never develop the condition it will be far in the future, and/or you got 'lucky'.
Re: (Score:2)
People who regularly eat fast food with similar heritage, sex, age and height weighs 16.3% more than the average of 193lbs at 5'3", that's interesting by itself.
193 pounds is average weight for a 5''3" person? That isn't interesting, it's terrifying. The normal/overweight border for that height is 141 pounds.
Re: (Score:2)
Never realized I was emaciated. Less than 160# at 6'2"....
Re: (Score:2)
and/or you got 'lucky'.
This is slashdot, people who live in their mom's basement don't get lucky you insensitive clod.
Re: (Score:1)
Do you know what you're talking about? What companies like 23andme are paid for is the sequencing. The amount of interpretation they include is very limited, partly due to regulations. The great thing is that you get your raw data and can share it on OpenHumans, or upload it to sites like Promethease and get it annotated, for free or almost free. What you get there is very transparent, with links to the studies it's based on. They don't even tell you "You are at risk", only that statistically for bearers of
It's not just the value of it (Score:2)
Re: (Score:2)
You're talking out your ass here, because this would be illegal as hell, and if an insurance company made a habit out o fit, they'd be out of business. Lab results are PHI, which means they are only provided to your doctor (or you if you request them.) Your insurance company can't deny based on something it doesn't know anything about. If the lab gave your results to your insurance company without your permission, they'd have committed a HIPAA violation. The same is true if your doctor gave them your result
Re: (Score:2)
It's not a swab. A swab is good enough for a genetic profile, i.e. what police use to match a sample of dna to a person, but it's not enough for an entire genome without PCR, which would raise the price quite a bit. You're supposed to spit a lot of times into a tube until you reach a fill line (this isn't that easy to do since you have to not eat or drink anything for 30 minutes prior to starting. (I had to rely on smelling smoked peppers to keep producing saliva.) Then you close the lid, which is full of a
Re: (Score:2)
There's a massive incentive to *find something* if you're paying to have a check done... if you get nothing reported then the value of the test seems to be nothing. If you are told to be "at risk" for some condition then you can go to your friends and tell them that you had no idea and that it was so worth it and that now you can take steps... and that perhaps they should get tested too. The best part is that if you're just reported to be "at risk", rather than an actual diagnosis, there's almost no need for accuracy. If you never develop the condition it will be far in the future, and/or you got 'lucky'.
Yep, just what I was thinking while I read the article. Additionally, high false positive rates, if not detected, eventually result in high intervention and prevention rates making the service look more effective than it actually is.
Re: (Score:2)
Re: (Score:2)
I spent 3 months of vacation in Canada and the USA once. Is that long enough for my DNA to have infused?
But seriously... I've heard that everyone who's an Anglo-Saxon pretty much is up to 10% Scandi (Viking raping and pillaging), 3% 'Jewish' and 2% Neatherthal. Beyond that it's possibly a waste of money.
Rational people (Score:2)
Understand that the only useful ancestral data are the YDNA and mtdna haplogroups, and that health data is 90% for the researchers.
It's rather natural when you think of probability (Score:2, Insightful)
Most "serious" genetic indicators are either quite rare, or their effect has already become apparent to consumer through other means. Common consumer genotyping tests test hundreds of thousands of SNPs. The rate of errors being at least 0.1-0.6% for these methods, there are bound to be hundreds of errors in a typical test result. People are not interested of benign errors, but are very interested on errors which would indicate a health-threatening condition, and this small fraction of test results gets lots
Holy biased data set, Batman! (Score:1)
By looking at people with worrying results, they are strongly filtering for people will results that are already likely to be false positives.
So that '40% false positive' number is nonsense.
Re: (Score:2)
Agreed. The original article is a massive statistics fail. You have to understand sensitivity vs. specificity. Without knowing the number of tests performed overall, you don't know anything. If they performed 10,000 tests, and those 46 people were the only ones with positives, and the result was incorrect on 40% of those (meaning 18 of them were actually negative despite the positive result) then the rate of false positives is 18/10000 = .18% which is great!
However, that's all pure speculation and BS th
Re: (Score:2)
Re: (Score:2)
It's not shit on this too much. Anyone who's had to work in any science related field, you're as good as your model and training data. To even make ancillary matches that maybe cause a bit of a 'concern' but later turn out to be benign or a false positive, um, sign me the fuck up? Especially if you're not in the health and welfare camp of "You don't know what you don't want to know". Time is on our side in improvement; of course we will, but I hope it's better for my children and not an ultimate target-on-the-back for pre-healthcare screening (which is already really is) and any future healthcare you plan to get. If accuracy is improved for the right reasons, absolutely, but humanity and greed tell me otherwise.
TFA is making it sound like we have world of mega hypochondriacs --- which probably isn't entirely out of the question with today's social media platform of "Oh everybody, on my way to another dicey check-up, fingers crossed " --- but by any measure, if there are genetic health concerns that run in my blood line, why not entertain it, especially if you want to be proactive and curb or neutralize it, if it's in your immediate control to change it? At least you had some concrete information as opposed to none.
We should shit on it more. "DTC companies" appear to be deliberately producing false results to sell more of their product. 40% is not a normal margin for error.
Re: (Score:2)
While I agree that DTC testing is dangerous and misleading, there is a bigger issue here... the numbers are all complete bunk because the way they selected these patients introduces a massive amount of bias. Given the way they described how they found the data, it is impossible to speak to true positives, true negatives, false positives, and false negatives. None of that can be known because they selected for patients who reported to their Dr. having a concerning previous result.
It's also worth noting tha
Re: (Score:2)
40% is not a normal margin for error.
Do you know that? I don't know details on this testing, but it sounds like 40% is exactly normal, as that's what we're seeing. If you're predicting the outcome of coin flips, a 40% error margin is pretty good. If you're determining whether a person is "at risk" for Alzheimer's, I can see the definition of "at risk" swaying the results dramatically.
Re: (Score:2)
40% is not a normal margin for error.
Do you know that? I don't know details on this testing, but it sounds like 40% is exactly normal, as that's what we're seeing. If you're predicting the outcome of coin flips, a 40% error margin is pretty good. If you're determining whether a person is "at risk" for Alzheimer's, I can see the definition of "at risk" swaying the results dramatically.
Yes, it's in TFA. Scientists at Ambry Genetics did actual testing and discovered the error. That's what the story is about.
Isn't that pretty good, if the conditions are rare (Score:3)
I remember an example from a book on probability: You have a test with a 1% false positive rate and a 1% false negative rate. It tests for a condition present in 1% of the population. 10000 people take your test. 100 of them actually have the condition, so you get 99 true positives from that; the 9900 people without the condition produce 99 false positives. By the criteria in the summary, in your population of people who sought further testing, 50% were false positives, but that doesn't make it a bad test.
Re: (Score:2)
No.
He is saying of the people that got a positive result, for certain genes, it was in fact a false positive. No information is provided about what percentage of people taking the test got a false negative or true negative. If a test is incredibly accurate at knowing if you are not at risk, that's still a valid test. Especially if it isn't too expensive or invasive to follow up on positive results to see if they were accurate.
You take the test, it says you're okay you go on with life.
You take the test, it
50%=the positive predicitve value (Score:2)
When 50% of your test positives actually have the condition that is a 50% positive predictive value, which is pretty good.
Re: (Score:2)
I wish I could give you mod points, but I've already commented on this thread... you are exactly right though... without knowing how many tests were performed you know nothing, and can't make a judgement about sensitivity and specificity.
Re: (Score:2)
I depends how the result of the test is used. If it jacks up insurance premiums then 50% false positives could be really bad. If it leads to a lifetime of worry it could be pretty terrible. But if all that happens is someone gets a more expensive, reliable test (cost amoritized over everyone who took the unreliable test) then it's probably not so bad.
Family Tree DNA? (Score:2)
Disclaimer - I have been testing my own and the DNA of others since 2007 with Family T
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
I've found at least one concrete discrepancy (Score:2)
Three years ago, I had my DNA analyzed by 23andme. According to my raw SNP data, there's no evidence of the genes for delayed sleep phase disorder... but I unquestionably have it, and have since childhood (& probably before). I found seemingly blatant discrepancies in a few other reported SNPs that appear to diverge quite a bit from observed reality.
I can think of a few possibilities:
* Promethase incorrectly interpreted my raw data & reported SNPs with incorrect values.
* The lab didn't actually sequ
Re: (Score:1)
* Your disorder isn't genetic
Most things aren't switches turned on or off by one gene.
Elizabeth Warren (Score:3)
All these political pundits keep pushing Elizabeth Warren to get a genetic test to "prove" she has Native American ancestry. They reference web-pages by these DNA testing co's in their claims that its possible to test. First, the vendors' own pages often state that the tests are imperfect, and that they often cannot rule out ethnic links. They are better able to say one is "probably" related to a given group, but poor at ruling out a relationship to a group. They look for specific markers or patterns, but currently not the entire genome.
Besides, Elizabeth going through with such tests is feeding the trolls. According her, her relatives told her when she was young that the family has at least one Native American in their background, and there's no reason to question one's older relatives about that. It's rude to the family, in my opinion, to publicly question their word. I hope she doesn't give in to the trolls the way Obama gave into T about the long-form birth certificate.
I see no reason to make a big deal about it. It's just partisan tiddly wings (which both sides do, by the way). Don't even get me started about T's rude "Pocahontas" jokes in front of a Native American event. Very tacky.
Re: (Score:1)
That doesn't contradict anything I've said.