Gene Genie
October 16, 2018 2:07 PM Subscribe
Most White Americans’ DNA Can Be Identified Through Genealogy Databases. "Already, 60 percent of Americans of Northern European descent — the primary group using these sites — can be identified through such databases whether or not they’ve joined one themselves, according to a study published today in the journal Science.
Within two or three years, 90 percent of Americans of European descent will be identifiable from their DNA, researchers found."
And thanks to my bro convincing both of our parents to do DNA tests then uploading all the results to various big data places, my lucrative future supervillain career is ruined. It was my retirement plan, dammit!
posted by scruss at 3:08 PM on October 16, 2018 [12 favorites]
posted by scruss at 3:08 PM on October 16, 2018 [12 favorites]
srboisvert, same except my genes include not giving a shit about my or anyone's ancestors. It also seems like if I did, the answers would be astonishingly boring, common and ultimately just lead to new ways to be exploited by the government and capitalists.
posted by GoblinHoney at 3:08 PM on October 16, 2018 [5 favorites]
posted by GoblinHoney at 3:08 PM on October 16, 2018 [5 favorites]
How long before you can just lick an Alexa for this information?
posted by grumpybear69 at 3:12 PM on October 16, 2018 [48 favorites]
posted by grumpybear69 at 3:12 PM on October 16, 2018 [48 favorites]
What's the recommended service for DNA ancestry these days? According to my cursory research, Ancestry.com is currently the price-leader at $69 (there's a sale). I have no idea how the services stack up though.
posted by w0mbat at 3:15 PM on October 16, 2018
posted by w0mbat at 3:15 PM on October 16, 2018
I want to say one word to you. Just one word.
Are you listening?
Metadata.
posted by seanmpuckett at 3:17 PM on October 16, 2018 [30 favorites]
Are you listening?
Metadata.
posted by seanmpuckett at 3:17 PM on October 16, 2018 [30 favorites]
In case you needed another reason to feel gross about these:
"In another example of the extraordinary lengths Canadian immigration officials go to deport migrants, the Canada Border Services Agency has been collecting their DNA and using ancestry websites to find and contact their distant relatives and establish their nationality."
posted by ITheCosmos at 3:18 PM on October 16, 2018 [12 favorites]
"In another example of the extraordinary lengths Canadian immigration officials go to deport migrants, the Canada Border Services Agency has been collecting their DNA and using ancestry websites to find and contact their distant relatives and establish their nationality."
posted by ITheCosmos at 3:18 PM on October 16, 2018 [12 favorites]
A friend's coworker (in the office next door to his) was recently arrested as a serial rapist because of GEDmatch. An interesting detail was that the police had a "John Doe arrest warrant" for someone matching a specific genetic profile. I find this so fascinating. Discussion on John Doe arrest warrants.
posted by vunder at 3:23 PM on October 16, 2018 [13 favorites]
posted by vunder at 3:23 PM on October 16, 2018 [13 favorites]
This is fabulous! Maybe we can get an accidental incest prevention app like they have in Iceland!
posted by heatherlogan at 3:31 PM on October 16, 2018 [8 favorites]
posted by heatherlogan at 3:31 PM on October 16, 2018 [8 favorites]
In the future, identical twins and triplets will be the new untouchable criminal overlords.
posted by aiglet at 3:31 PM on October 16, 2018 [11 favorites]
posted by aiglet at 3:31 PM on October 16, 2018 [11 favorites]
In all the dystopian films coming to mind at the moment, your introduction and the bulk of the film is in the dystopia. You never quite learn how society got there on screen in a lengthy fashion. Is this what it feels like? One day seeing this, next day seeing the latest Boston Dynamics video, etc?
posted by cashman at 3:41 PM on October 16, 2018 [19 favorites]
posted by cashman at 3:41 PM on October 16, 2018 [19 favorites]
GEDmatch totally sounds like a dating site for high school dropouts (like me).
A few years ago, UC Berkeley issued DNA testing kits to all of their incoming students with "anonymized" results. Bless 'em, a majority of the students declined, saying it was creepy and invasive. Interesting to know that those "anonymized" results could still be used to target someone. That's just grand.
posted by shapes that haunt the dusk at 3:45 PM on October 16, 2018 [18 favorites]
A few years ago, UC Berkeley issued DNA testing kits to all of their incoming students with "anonymized" results. Bless 'em, a majority of the students declined, saying it was creepy and invasive. Interesting to know that those "anonymized" results could still be used to target someone. That's just grand.
posted by shapes that haunt the dusk at 3:45 PM on October 16, 2018 [18 favorites]
You had law enforcement accessing data from an open source site, uploading data and getting matches like anybody else. The only difference was they were using the matches for investigatory purposes.But that one difference makes all the difference. We need laws around use of genetic information, because the risk and impact of abuse is too great.
posted by NoxAeternum at 3:46 PM on October 16, 2018 [6 favorites]
23&me and Ancestry.com make me so angry. DNA are not not culture. Genetics are not ethnicity. And genetic databases from these organizations should not be commodities for police, or for proprietary medical companies to use for profit. It is so unethical and it fuels pseudoscientific racism and nonsense like Elizabeth Warren's DNA tests. I can't wait until the next time I teach Introduction to Human Evolution so that I can dig into this with students. Until then, check out Jennifer Raff, Kim TallBear, Rick Smith, Carl Zimmer, and others.
posted by ChuraChura at 4:48 PM on October 16, 2018 [38 favorites]
posted by ChuraChura at 4:48 PM on October 16, 2018 [38 favorites]
Yeah, what ChuraChura said. The whole basis for these tests (as tools for tracing geographic or ethnic roots) is reductive and deeply flawed. I would heartily recommend Kim TallBear's book Native American DNA for a thorough debunking.
posted by shapes that haunt the dusk at 4:58 PM on October 16, 2018 [12 favorites]
posted by shapes that haunt the dusk at 4:58 PM on October 16, 2018 [12 favorites]
More important than privacy or preventing dystopia or anything like that, I think we need a mascot for accidental incest prevention. Like McGruff the Crime Dog or Smokey the Bear. Charles II the Prophylaxis Bonobo?
posted by XMLicious at 5:08 PM on October 16, 2018 [4 favorites]
posted by XMLicious at 5:08 PM on October 16, 2018 [4 favorites]
Yeah, what ChuraChura said. The whole basis for these tests (as tools for tracing geographic or ethnic roots) is reductive and deeply flawed.
I heard something on NPR (or at least the local NPR affiliate) a month or two ago about precisely this topic. Everybody knows that the whole "tell your genetic story!" thing is mostly bullshit. But, see, after it became possible to sequence a person's DNA without a Manhattan Project style investment, scientists needed access to a lot of people's DNA in order to do their research. They started paying people for samples of their DNA. But people were suspicious and weren't signing up all that fast. So that way was slow and expensive.
Then some genius realized that, while they couldn't get what they needed by paying people, they could market better and get people to instead PAY THE SCIENTISTS for the "privilege" of having their DNA sequenced as long as they provided a mostly bullshit geographical and historical analysis. Yes, scientists couldn't get enough samples when they were paying for them so they instead convinced people to send in DNA themselves.. along with cash and a thank you.
Brilliant. And evil. But mostly brilliant.
posted by Justinian at 5:39 PM on October 16, 2018 [18 favorites]
I heard something on NPR (or at least the local NPR affiliate) a month or two ago about precisely this topic. Everybody knows that the whole "tell your genetic story!" thing is mostly bullshit. But, see, after it became possible to sequence a person's DNA without a Manhattan Project style investment, scientists needed access to a lot of people's DNA in order to do their research. They started paying people for samples of their DNA. But people were suspicious and weren't signing up all that fast. So that way was slow and expensive.
Then some genius realized that, while they couldn't get what they needed by paying people, they could market better and get people to instead PAY THE SCIENTISTS for the "privilege" of having their DNA sequenced as long as they provided a mostly bullshit geographical and historical analysis. Yes, scientists couldn't get enough samples when they were paying for them so they instead convinced people to send in DNA themselves.. along with cash and a thank you.
Brilliant. And evil. But mostly brilliant.
posted by Justinian at 5:39 PM on October 16, 2018 [18 favorites]
A lot of my students are involved in a student organization, run by a genetics professor and affiliated with the Biology department, that educates the general public about the uses and limitations of commercial DNA tests. I like the idea, but they all take a commercial DNA test when they join the organization, and I want to tell them to think very, very, very carefully about that before they do it. (The organization got some sort of grant to fund the students' DNA tests. I don't know if it's required, but nobody has told me that they decided not to do it.) I really worry about the implications of voluntarily giving your DNA information to commercial entities that are not bound by medical privacy laws. And I'm not sure that I think that the average 18-year-old has really thought through those implications, nor does Dr. Fancy Geneticist seem all that interested in those issues. My sense is that he's very interested in the science, and he's a little interested in things like the way that race and ethnicity don't overlap with anything that can be determined by a genetic test, but he's not real interested in the complicated ways that DNA testing can play out in your actual life. So they do presentations on how your genetics doesn't determine your ethnicity, but they haven't done anything on "keep in mind that you may find out that your father isn't biologically your father" or "have you considered that your sisters, daughters, and nieces could potentially face discrimination because you test positive for BRCA?"
Personally, I'm wrestling with the whole question of whether I should get BRCA testing, which would be done in a medical setting, with relevant medical privacy laws in force, and I'm still a little freaked out about having that information out there in the universe. Right now, there are laws in the US that prevent things like employment discrimination on the basis of genetic tests, but given the state of the world, I am not confident that will continue to be the case forever.
posted by ArbitraryAndCapricious at 5:40 PM on October 16, 2018 [7 favorites]
Personally, I'm wrestling with the whole question of whether I should get BRCA testing, which would be done in a medical setting, with relevant medical privacy laws in force, and I'm still a little freaked out about having that information out there in the universe. Right now, there are laws in the US that prevent things like employment discrimination on the basis of genetic tests, but given the state of the world, I am not confident that will continue to be the case forever.
posted by ArbitraryAndCapricious at 5:40 PM on October 16, 2018 [7 favorites]
Until then, check out Jennifer Raff, Kim TallBear, Rick Smith, Carl Zimmer, and others.
For podcast-listening types, Kim TallBear was recently part of a great Media Indigena podcast panel discussion on the topic:
Ep. 119: Why DNA "Indigenous ancestry" tests ain't worth a doggone dime
Then some genius realized that, while they couldn't get what they needed by paying people, they could market better and get people to instead PAY THE SCIENTISTS for the "privilege" of having their DNA sequenced as long as they provided a mostly bullshit geographical and historical analysis.
Heh. Sawbones also recently did a show on these other problems with DNA testing services.
posted by mandolin conspiracy at 5:49 PM on October 16, 2018 [5 favorites]
For podcast-listening types, Kim TallBear was recently part of a great Media Indigena podcast panel discussion on the topic:
Ep. 119: Why DNA "Indigenous ancestry" tests ain't worth a doggone dime
Then some genius realized that, while they couldn't get what they needed by paying people, they could market better and get people to instead PAY THE SCIENTISTS for the "privilege" of having their DNA sequenced as long as they provided a mostly bullshit geographical and historical analysis.
Heh. Sawbones also recently did a show on these other problems with DNA testing services.
posted by mandolin conspiracy at 5:49 PM on October 16, 2018 [5 favorites]
Once again, trashy cousins ruin it for everyone.
posted by Hypatia at 7:18 PM on October 16, 2018 [3 favorites]
posted by Hypatia at 7:18 PM on October 16, 2018 [3 favorites]
So far they've gotten at least one alleged serial rapist via GEDMatch, there's at least one other high profile case being worked similarly, and DNA testing in general has led to a significant number of cold-case closures, as well as the exoneration of hundreds from death row. While certainly not perfect—and both the public and law enforcement (and especially the judiciary) need education on the limitations of DNA sequencing and matching—it seems cautiously like a net positive to me.
I'm not sure that having private companies run the databases is exactly, uh, awesome, but naturally this is America and I can only imagine the simultaneous hue and cry from both the left and right if it was suggested that the FBI do it (which seems like the logical choice, since they already maintain the largest national fingerprint database and some other biometrics), so that gets filed under Nice Things We Can't Have, at least for the foreseeable future.
But in all seriousness, the NorCal Rapist thing makes me quite tempted to upload my genome to GEDMatch, just on the off chance that some distant relative of mine happens to be a monster. Not that I have reason to think that, of course, but presumably neither did Roy Charles Waller's relatives, and look what that turned up. I have exactly zero compunction about sending some genetically-related person up the river if it turns out they've got a John Doe warrant open against them, particularly considering the type of crimes that are likely to result in a John Doe warrant for DNA (rape by an unknown assailant being the most common, which tends to be a repeated crime, and one with a terribly low resolution rate prior to DNA evidence).
Certainly room for regulation of the databases (making them strictly liable for damages suffered as the result of a data breach would be an interesting starting point), and for rules of evidence covering "matches" (which tend to be probabilities rather than Boolean YES|NOs, something that I think is not clear in the public's mind and therefore ripe for abuse in jury trials), but I was somewhat... surprised... when the NorCal Rapist case was all over the news, how many people would casually seem to put themselves in the position of the perpetrator, or were strangely judgey of the unnamed distant relatives who uploaded their DNA and contributed to catching the guy.
I guess maybe I shouldn't be quite so surprised, given other events over the last couple of years, that so many people around seem to side with probable serial rapists given the opportunity, but... jeez. Those conversations were depressing.
posted by Kadin2048 at 8:05 PM on October 16, 2018 [7 favorites]
I'm not sure that having private companies run the databases is exactly, uh, awesome, but naturally this is America and I can only imagine the simultaneous hue and cry from both the left and right if it was suggested that the FBI do it (which seems like the logical choice, since they already maintain the largest national fingerprint database and some other biometrics), so that gets filed under Nice Things We Can't Have, at least for the foreseeable future.
But in all seriousness, the NorCal Rapist thing makes me quite tempted to upload my genome to GEDMatch, just on the off chance that some distant relative of mine happens to be a monster. Not that I have reason to think that, of course, but presumably neither did Roy Charles Waller's relatives, and look what that turned up. I have exactly zero compunction about sending some genetically-related person up the river if it turns out they've got a John Doe warrant open against them, particularly considering the type of crimes that are likely to result in a John Doe warrant for DNA (rape by an unknown assailant being the most common, which tends to be a repeated crime, and one with a terribly low resolution rate prior to DNA evidence).
Certainly room for regulation of the databases (making them strictly liable for damages suffered as the result of a data breach would be an interesting starting point), and for rules of evidence covering "matches" (which tend to be probabilities rather than Boolean YES|NOs, something that I think is not clear in the public's mind and therefore ripe for abuse in jury trials), but I was somewhat... surprised... when the NorCal Rapist case was all over the news, how many people would casually seem to put themselves in the position of the perpetrator, or were strangely judgey of the unnamed distant relatives who uploaded their DNA and contributed to catching the guy.
I guess maybe I shouldn't be quite so surprised, given other events over the last couple of years, that so many people around seem to side with probable serial rapists given the opportunity, but... jeez. Those conversations were depressing.
posted by Kadin2048 at 8:05 PM on October 16, 2018 [7 favorites]
I m also opting out, but imagine that, soon this will be socially mandatory, like how I have to be on facebook for the "communication s"part of my job
I imagine this will get worse until there will be direct consequences for not participating
posted by eustatic at 8:15 PM on October 16, 2018 [2 favorites]
I imagine this will get worse until there will be direct consequences for not participating
posted by eustatic at 8:15 PM on October 16, 2018 [2 favorites]
Speaking of DNA analysis and relationships, this awful news story in the Sydney Morning Herald has it all:
1. incomplete (and thus biased) datasets resulting in what's quite likely to be a false positive
2. Australian authorities arbitrarily applying a test to people of colour that they almost certainly wouldn't apply to white migrants
3. a Department claiming to have 'grounds for suspicion' but not explaining how they arrived at those grounds
4. forced family separation
5. a bureaucratic inflexibility that in no way contrasts with the Minister for Home Affairs seeming perfectly happy to intervene when the migrant in question is white and works for friends of his
6. for an extra little soupçon of Hellworld, the gig economy
posted by MarchHare at 8:51 PM on October 16, 2018 [4 favorites]
1. incomplete (and thus biased) datasets resulting in what's quite likely to be a false positive
2. Australian authorities arbitrarily applying a test to people of colour that they almost certainly wouldn't apply to white migrants
3. a Department claiming to have 'grounds for suspicion' but not explaining how they arrived at those grounds
4. forced family separation
5. a bureaucratic inflexibility that in no way contrasts with the Minister for Home Affairs seeming perfectly happy to intervene when the migrant in question is white and works for friends of his
6. for an extra little soupçon of Hellworld, the gig economy
posted by MarchHare at 8:51 PM on October 16, 2018 [4 favorites]
People from some ethnic groups are going to get a lot of false positives because they come from a smallish gene pool. For instance, I was trying to trace one person's ancestry via DNA when other sources had failed and I was absolutely stymied by the ridiculous number of third-cousins they appeared to have. I've heard anecdotally that all Ashkenazi Jews are apparently related at a fourth- or fifth-cousin level; I bet some other groups are even tighter than that.
posted by Joe in Australia at 9:44 PM on October 16, 2018 [4 favorites]
posted by Joe in Australia at 9:44 PM on October 16, 2018 [4 favorites]
Law enforcement searches for the truth until they find a suspect, and then they search for evidence to convict. DNA evidence has been powerful enough to show how often they're wrong and how many innocent people have been punished. Its power can also be used to cast suspicion with no other basis, and likely outweigh any reasonable amount of evidence the suspect could supply in a decades-old case.
Even with good intentions, DNA contamination has led to false matches. If the primary investigative tool is DNA, even a billion-to-one chance is likely to snare innocent people. This is because you aren't identifying the one-in-a-billion, you're making millions of tests with a billion-to-one chance of being wrong each time. DNA tests are not a genome match that narrows it down to identical twins: they are approximate, and it's likely no further tests on behalf of the defense are possible due to evidence being consumed.
The other issue is why are we doing this. There are supposed to be three benefits to criminal justice: rehabilitation, prevention, and solace to victims. There's no rehabilitation in convicting someone that's been law-abiding for decades, and likely not a lot of benefit to victims. Of course the message that you can't ever escape your crimes is pretty powerful, but I think if this sort of case is ever proven false that's all out the window.
posted by netowl at 11:27 PM on October 16, 2018 [9 favorites]
Even with good intentions, DNA contamination has led to false matches. If the primary investigative tool is DNA, even a billion-to-one chance is likely to snare innocent people. This is because you aren't identifying the one-in-a-billion, you're making millions of tests with a billion-to-one chance of being wrong each time. DNA tests are not a genome match that narrows it down to identical twins: they are approximate, and it's likely no further tests on behalf of the defense are possible due to evidence being consumed.
The other issue is why are we doing this. There are supposed to be three benefits to criminal justice: rehabilitation, prevention, and solace to victims. There's no rehabilitation in convicting someone that's been law-abiding for decades, and likely not a lot of benefit to victims. Of course the message that you can't ever escape your crimes is pretty powerful, but I think if this sort of case is ever proven false that's all out the window.
posted by netowl at 11:27 PM on October 16, 2018 [9 favorites]
The other issue is why are we doing this
It's true that transformative technology often has unforeseen consequences, but it also sometimes has tremendous benefits. For example, it could reunite long lost twins. Or, hypothetically, it could help to connect a recent spree of sloppy and sexually bizarre burglaries to a cheek swab mailed to 23AndMe under the name "Ted Cruz". We just have to take the good with the bad.
posted by qxntpqbbbqxl at 11:40 PM on October 16, 2018 [4 favorites]
It's true that transformative technology often has unforeseen consequences, but it also sometimes has tremendous benefits. For example, it could reunite long lost twins. Or, hypothetically, it could help to connect a recent spree of sloppy and sexually bizarre burglaries to a cheek swab mailed to 23AndMe under the name "Ted Cruz". We just have to take the good with the bad.
posted by qxntpqbbbqxl at 11:40 PM on October 16, 2018 [4 favorites]
I mean, yeah, you can always catch more criminals if you're willing to ignore the general public's expectations of privacy.
posted by shapes that haunt the dusk at 12:58 AM on October 17, 2018 [6 favorites]
posted by shapes that haunt the dusk at 12:58 AM on October 17, 2018 [6 favorites]
I am a little curious about a family background secret story which might be resolved by one of these services, but there's no way I'm handing my DNA over to some corporation to satisfy that curiousity.
posted by pompomtom at 3:00 AM on October 17, 2018 [1 favorite]
posted by pompomtom at 3:00 AM on October 17, 2018 [1 favorite]
And then the testing company gets bought out and suddenly you can't get health insurance or a home loan.
posted by ambulocetus at 4:22 AM on October 17, 2018 [5 favorites]
posted by ambulocetus at 4:22 AM on October 17, 2018 [5 favorites]
There's no rehabilitation in convicting someone that's been law-abiding for decades
Hasn’t been caught is not the same as law-abiding.
posted by snofoam at 5:20 AM on October 17, 2018 [3 favorites]
Hasn’t been caught is not the same as law-abiding.
posted by snofoam at 5:20 AM on October 17, 2018 [3 favorites]
Two of my family paid for these tests. They came back (in agreement) saying that we are one-eighth Native American. I suppose that is reasonable: the Latino side of my family has been in New Mexico since the 1630s. I want it to be true.
posted by dances_with_sneetches at 6:11 AM on October 17, 2018
posted by dances_with_sneetches at 6:11 AM on October 17, 2018
My mother's cousins (all sixty-two million of them) are very much into the genealogy/family history part of this. The family trees; the who went to which county when and who crossed which ocean/national border when, and how many people are named after the same uncle.
Basically, there was no hope of me staying out of these databases, no matter how much I wanted to, once my mother's giant old-school Catholic families got involved. So many of them sent in their samples and then spent days entering all the information they had about who was related to them. While they may not have my spit on a stick, but they've got everything else possible about me. It's depressing.
posted by crush at 6:41 AM on October 17, 2018
Basically, there was no hope of me staying out of these databases, no matter how much I wanted to, once my mother's giant old-school Catholic families got involved. So many of them sent in their samples and then spent days entering all the information they had about who was related to them. While they may not have my spit on a stick, but they've got everything else possible about me. It's depressing.
posted by crush at 6:41 AM on October 17, 2018
Mod note: One deleted. The problem with untagged sarcasm / ironic voice / "I'm saying what [bad people] would say to show it's bad" comments is that they are almost never clear enough to avoid confusion, anger, and ensuing needless arguments and explanations, so better to either just be clear, or super explicit that you are not actually putting forth that faux argument. Thanks.
posted by taz (staff) at 6:42 AM on October 17, 2018 [2 favorites]
posted by taz (staff) at 6:42 AM on October 17, 2018 [2 favorites]
Two of my family paid for these tests. They came back (in agreement) saying that we are one-eighth Native American. I suppose that is reasonable: the Latino side of my family has been in New Mexico since the 1630s. I want it to be true.
posted by dances_with_sneetches at 8:11 AM on October 17
No sneetch though?
posted by srboisvert at 8:01 AM on October 17, 2018 [1 favorite]
posted by dances_with_sneetches at 8:11 AM on October 17
No sneetch though?
posted by srboisvert at 8:01 AM on October 17, 2018 [1 favorite]
taz: I wasn't being ironic. It's a tool, and there may be a moral obligation to use it as part of being a good citizen. Same as recycling or voting or reporting crimes.
posted by Leon at 8:11 AM on October 17, 2018
posted by Leon at 8:11 AM on October 17, 2018
And then the testing company gets bought out and suddenly you can't get health insurance or a home loan.
People should be required to watch Gattaca before submitting to these tests.
posted by dirigibleman at 8:27 AM on October 17, 2018 [2 favorites]
People should be required to watch Gattaca before submitting to these tests.
posted by dirigibleman at 8:27 AM on October 17, 2018 [2 favorites]
Star-bellies are overrated.
posted by dances_with_sneetches at 8:34 AM on October 17, 2018 [1 favorite]
posted by dances_with_sneetches at 8:34 AM on October 17, 2018 [1 favorite]
The other issue is why are we doing this. There are supposed to be three benefits to criminal justice: rehabilitation, prevention, and solace to victims. There's no rehabilitation in convicting someone that's been law-abiding for decades, and likely not a lot of benefit to victims. Of course the message that you can't ever escape your crimes is pretty powerful, but I think if this sort of case is ever proven false that's all out the window.
We're talking here about serial rapists and murderers, I think there's a lot of solace to victims and families knowing that the person has been caught. The victims of the Norcal rapist expressed as much. I mean, that guy committed those rapes over 15 years, the last known one was 12 years ago. It's not like this is 50 years ago. Also, this technology could stop a serial rapist/murderer in a much more timely fashion, preventing future crimes - the long timelines are only because the technology is new. Statutes of limitations come into play for less serious crimes.
I don't discount the dangers of this type of technology and information, but your "why do this?" line of reasoning is bogus when the actual cases that have been worked are taken into consideration. I think if you look at it from the perspective of the victims, you'd have trouble shrugging your shoulders and saying "why?" particularly if you read some of the details. (The Norcal rapist attacked women in their homes and held them captive for hours. One of his victims he called on the phone at her work less than 3 weeks later to apologize. Can you imagine?)
posted by vunder at 9:58 AM on October 17, 2018 [4 favorites]
We're talking here about serial rapists and murderers, I think there's a lot of solace to victims and families knowing that the person has been caught. The victims of the Norcal rapist expressed as much. I mean, that guy committed those rapes over 15 years, the last known one was 12 years ago. It's not like this is 50 years ago. Also, this technology could stop a serial rapist/murderer in a much more timely fashion, preventing future crimes - the long timelines are only because the technology is new. Statutes of limitations come into play for less serious crimes.
I don't discount the dangers of this type of technology and information, but your "why do this?" line of reasoning is bogus when the actual cases that have been worked are taken into consideration. I think if you look at it from the perspective of the victims, you'd have trouble shrugging your shoulders and saying "why?" particularly if you read some of the details. (The Norcal rapist attacked women in their homes and held them captive for hours. One of his victims he called on the phone at her work less than 3 weeks later to apologize. Can you imagine?)
posted by vunder at 9:58 AM on October 17, 2018 [4 favorites]
I received one of these as a Christmas gift, and promptly passed it along elsewhere after I wasn't able to return the kit. I found it creepy and invasive, and was not encouraged when a friend, who works in pharmaceutical research, mentioned how her company was buying the results of data from these types of companies for their research.
posted by PearlRose at 10:00 AM on October 17, 2018 [1 favorite]
posted by PearlRose at 10:00 AM on October 17, 2018 [1 favorite]
I've heard anecdotally that all Ashkenazi Jews are apparently related at a fourth- or fifth-cousin level;
Yes, my husband and his childhood friend discovered they were cousins after his mom became a genealogy buff.
posted by bq at 11:42 AM on October 17, 2018
Yes, my husband and his childhood friend discovered they were cousins after his mom became a genealogy buff.
posted by bq at 11:42 AM on October 17, 2018
People from some ethnic groups are going to get a lot of false positives because they come from a smallish gene pool.
Those aren't actually "false positives", though; the effect of endogamy is to make very distant relationships appear closer than they actually are (since it magnifies the chances of any particular DNA segment surviving recombination). So those third or fourth cousins may actually be sixth or seventh or eighth cousins (and probably 6th or 7th or 8th cousins through multiple shared ancestors).
posted by Pseudonymous Cognomen at 12:17 PM on October 17, 2018 [2 favorites]
Those aren't actually "false positives", though; the effect of endogamy is to make very distant relationships appear closer than they actually are (since it magnifies the chances of any particular DNA segment surviving recombination). So those third or fourth cousins may actually be sixth or seventh or eighth cousins (and probably 6th or 7th or 8th cousins through multiple shared ancestors).
posted by Pseudonymous Cognomen at 12:17 PM on October 17, 2018 [2 favorites]
I do kind of wonder if there is some level of generational endpoint of getting these tests done, though. If both my parents take them, what's the point for me or my siblings to take the tests? (I guess if we thought we weren't related, but based on appearance, no worries there). I suppose if your folks never share their results with you, then you may want to get tested later, but part of the marketing around the testing to white people is to shout from the rooftops your ancestral roots. I could see a whole generation of baby boomers doing the tests, and then a generation of gen x/millenials going "eh!" but maybe by that point they've sequenced enough white genes in the US.
I will say that as a generic mutt american, going back through the genealogy trees and seeing how my family is part of the genocidal westward expansion of America is both enlightening and sad. At some point I think I need to read up on incorporating genealogy with reparations.
posted by Hermeowne Grangepurr at 12:35 PM on October 17, 2018
I will say that as a generic mutt american, going back through the genealogy trees and seeing how my family is part of the genocidal westward expansion of America is both enlightening and sad. At some point I think I need to read up on incorporating genealogy with reparations.
posted by Hermeowne Grangepurr at 12:35 PM on October 17, 2018
Earlier this year I had a heated argument with a woman of Guyanese background about this. She had been speaking wistfully of her ancestry, and how she wanted to investigate it through genetic testing. I told her it was a bad idea, because of the commercial nature of these companies, the lack of medical privacy law, and the potential for abuse.
One comment she had made stuck with me. She laughed when I said that the data could be used for racist purposes, "because I'm already visibly brown." My retort was that although you are visibly brown, they wouldn't need to meet you in person, have you register with a government agency, submit a census report, or go through any other process of documentation to discover the colour of your skin and reveal any segments of your ancestry that they might consider relevant for the purposes of discrimination.
The Nazis had to appropriate census data, church records; marriage records, baptismal records to support their abominable racial laws. If commercial genetic testing becomes widespread, that data will make new, more granular types of discrimination possible. You could choose to round up everyone of x% (above the margin of error) y heritage. Like someone else pointed out, you could choose to deny insurance based on the presence of specific genes. And if this information is commercially available, what stops a bunch of alt righter youths from chipping in together, making a front corporation, and then buying this data to terrorize individuals?
posted by constantinescharity at 12:53 PM on October 17, 2018 [2 favorites]
One comment she had made stuck with me. She laughed when I said that the data could be used for racist purposes, "because I'm already visibly brown." My retort was that although you are visibly brown, they wouldn't need to meet you in person, have you register with a government agency, submit a census report, or go through any other process of documentation to discover the colour of your skin and reveal any segments of your ancestry that they might consider relevant for the purposes of discrimination.
The Nazis had to appropriate census data, church records; marriage records, baptismal records to support their abominable racial laws. If commercial genetic testing becomes widespread, that data will make new, more granular types of discrimination possible. You could choose to round up everyone of x% (above the margin of error) y heritage. Like someone else pointed out, you could choose to deny insurance based on the presence of specific genes. And if this information is commercially available, what stops a bunch of alt righter youths from chipping in together, making a front corporation, and then buying this data to terrorize individuals?
posted by constantinescharity at 12:53 PM on October 17, 2018 [2 favorites]
The argument for going after cold-case serial rapists and murderers—even if we're pretty sure that they're not raping or murdering anymore—is the same as hunting and prosecuting Auschwitz camp guards even when they've been living in Illinois under an assumed name for decades. It's to prove the point that our society is stronger than their individual evil, and that even if the arc of the moral universe is very long indeed, it bends toward justice; you can't do evil and then just recede back into the woodwork to live amongst the rest of us. The idea that you can get away with crime, especially crime as damaging and impossible to correct as rape or murder, is socially corrosive; it is anathema to the very idea of justice itself.
That said, there are legitimate concerns with false positives if you're searching through a database of millions (potentially billions) of people and have even a very slight error rate. This, IMO anyway, is something that we need to address by educating the public and the judiciary on how these tools work, that they are by their nature imperfect and not a crystal ball, not by refusing to use them where they are appropriate.
posted by Kadin2048 at 1:37 PM on October 17, 2018 [4 favorites]
That said, there are legitimate concerns with false positives if you're searching through a database of millions (potentially billions) of people and have even a very slight error rate. This, IMO anyway, is something that we need to address by educating the public and the judiciary on how these tools work, that they are by their nature imperfect and not a crystal ball, not by refusing to use them where they are appropriate.
posted by Kadin2048 at 1:37 PM on October 17, 2018 [4 favorites]
I've avoided any commercial/non-medical DNA tests, mostly out of a mix of (1) wanting to avoid false positives for any medical issues, (2) privacy concerns, and (3) and my family background is really not mysterious.
But Sloan Kettering is selling my tissue samples to a medical start up. And I elected to do both carrier screenings and NIPT testing during this pregnancy (despite being slightly annoyed at my ob gyn for not being able to give me good answers about privacy protection and false positive rates). So now I'm left with trusting those companies to follow the laws they have to follow regarding patient privacy, which, um, yeah.
posted by damayanti at 1:48 PM on October 17, 2018
But Sloan Kettering is selling my tissue samples to a medical start up. And I elected to do both carrier screenings and NIPT testing during this pregnancy (despite being slightly annoyed at my ob gyn for not being able to give me good answers about privacy protection and false positive rates). So now I'm left with trusting those companies to follow the laws they have to follow regarding patient privacy, which, um, yeah.
posted by damayanti at 1:48 PM on October 17, 2018
There are supposed to be three benefits to criminal justice: rehabilitation, prevention, and solace to victims.
There's a big fourth one: accountability.
posted by storybored at 9:44 PM on October 17, 2018
There's a big fourth one: accountability.
posted by storybored at 9:44 PM on October 17, 2018
the effect of endogamy is to make very distant relationships appear closer than they actually are (since it magnifies the chances of any particular DNA segment surviving recombination). So those third or fourth cousins may actually be sixth or seventh or eighth cousins (and probably 6th or 7th or 8th cousins through multiple shared ancestors).
Nobody is interested in sixth cousins though, not even law enforcement. And if there are sixth cousins being read as third cousins, I presume you're more likely to have third cousins being read as siblings. DNA evidence has an aura of infallibility around it and I'm not sure juries will be able to properly assess an argument that the accused comes from a restricted gene pool - especially if that ethnic group hasn't been well studied and there's no expert testimony to that effect.
posted by Joe in Australia at 11:21 PM on October 17, 2018
Nobody is interested in sixth cousins though, not even law enforcement. And if there are sixth cousins being read as third cousins, I presume you're more likely to have third cousins being read as siblings. DNA evidence has an aura of infallibility around it and I'm not sure juries will be able to properly assess an argument that the accused comes from a restricted gene pool - especially if that ethnic group hasn't been well studied and there's no expert testimony to that effect.
posted by Joe in Australia at 11:21 PM on October 17, 2018
*not a comment on you Joe, just using this as a jumping of point*
And if there are sixth cousins being read as third cousins, I presume you're more likely to have third cousins being read as siblings.
Gah all of this is because of a lack of understanding of genes, science, and probabilities.
First, there's nothing in the DNA that says "3rd Cousin". We have decided on how much DNA should be shared between different levels of family members. And as soon as science draws a hard line around something, there's always nature to come around and spit in it's face.
Second, all this DNA stuff is probabilities. And probabilities are too often confused with certainties. All humans share, what, 99.9% of genes or something. And that left over 0.1% is what is used for DNA testing, except they're not testing 100% of that 0.1%, they're testing a portion that, mathematically, has been deemed sufficient to produce 1-in-a-billion results, or whatever you're aiming for.
If someone told me that, through DNA testing, I was related to someone with a one-in-a-billion probability of error, I would believe it. However, I would also assume that if that same test is done, say, 7 billion times, I would expect there to be 7 false positives. Maybe more, maybe less. (Being an actuary, I live with the fact that I predict the future and it will be wrong. The tricky part is trying to guess by how much). I guess really the only way to reduce false positives/negatives from sampling error would be to compare the entire 0.1% that is unique. Then the errors arise from the procedure itself, and not by chance.
DNA is a tool, not an answer, and it should be treated with appropriate skepticism. But humans suck a nuance so this is the situation we're in.
(Also, as an actuary, I am in love with data science, and I think to be able to sequence like an entire generation would be so fucking cool. But the privacy concerns are way too serious for that.)
posted by LizBoBiz at 1:17 AM on October 18, 2018 [3 favorites]
And if there are sixth cousins being read as third cousins, I presume you're more likely to have third cousins being read as siblings.
Gah all of this is because of a lack of understanding of genes, science, and probabilities.
First, there's nothing in the DNA that says "3rd Cousin". We have decided on how much DNA should be shared between different levels of family members. And as soon as science draws a hard line around something, there's always nature to come around and spit in it's face.
Second, all this DNA stuff is probabilities. And probabilities are too often confused with certainties. All humans share, what, 99.9% of genes or something. And that left over 0.1% is what is used for DNA testing, except they're not testing 100% of that 0.1%, they're testing a portion that, mathematically, has been deemed sufficient to produce 1-in-a-billion results, or whatever you're aiming for.
If someone told me that, through DNA testing, I was related to someone with a one-in-a-billion probability of error, I would believe it. However, I would also assume that if that same test is done, say, 7 billion times, I would expect there to be 7 false positives. Maybe more, maybe less. (Being an actuary, I live with the fact that I predict the future and it will be wrong. The tricky part is trying to guess by how much). I guess really the only way to reduce false positives/negatives from sampling error would be to compare the entire 0.1% that is unique. Then the errors arise from the procedure itself, and not by chance.
DNA is a tool, not an answer, and it should be treated with appropriate skepticism. But humans suck a nuance so this is the situation we're in.
(Also, as an actuary, I am in love with data science, and I think to be able to sequence like an entire generation would be so fucking cool. But the privacy concerns are way too serious for that.)
posted by LizBoBiz at 1:17 AM on October 18, 2018 [3 favorites]
if there are sixth cousins being read as third cousins, I presume you're more likely to have third cousins being read as siblings
That isn't going to happen, unless your family tree looks like the Spanish Habsburgs'. As noted above, the DNA companies, relationship labels are there because there are average amounts of DNA people share at different relationship levels; those averages have been established based on unrelated people in the broad general population. The elevated amounts of shared DNA observed in Ashkenazi Jews (and in some other populations, like Quebecois, and Mennonites) are the result of a small founding population and centuries of intermarriage between the descendants of that small founding population.
I've used my parents' DNA test results for genealogy; my mother (whose ancestors include English Catholics who emigrated to Maryland in the 17th century) shows similar effects of endogamy (as that founding population of Catholic Marylanders intermarried for the next hundred years or so, and then their descendants went to Kentucky, and continued intermarrying). She has a lot of estimated "4th cousins" who are actually 7th or 8th cousins, but share two dozen different ancestors from that small founding population.
Because of the way DNA recombination works, the only case where you'll have (first) cousins being read as (half) siblings is for the children of identical twins (who are, genetically speaking, essentially the same person).
posted by Pseudonymous Cognomen at 9:58 AM on October 18, 2018
That isn't going to happen, unless your family tree looks like the Spanish Habsburgs'. As noted above, the DNA companies, relationship labels are there because there are average amounts of DNA people share at different relationship levels; those averages have been established based on unrelated people in the broad general population. The elevated amounts of shared DNA observed in Ashkenazi Jews (and in some other populations, like Quebecois, and Mennonites) are the result of a small founding population and centuries of intermarriage between the descendants of that small founding population.
I've used my parents' DNA test results for genealogy; my mother (whose ancestors include English Catholics who emigrated to Maryland in the 17th century) shows similar effects of endogamy (as that founding population of Catholic Marylanders intermarried for the next hundred years or so, and then their descendants went to Kentucky, and continued intermarrying). She has a lot of estimated "4th cousins" who are actually 7th or 8th cousins, but share two dozen different ancestors from that small founding population.
Because of the way DNA recombination works, the only case where you'll have (first) cousins being read as (half) siblings is for the children of identical twins (who are, genetically speaking, essentially the same person).
posted by Pseudonymous Cognomen at 9:58 AM on October 18, 2018
Why White Supremacists Are Chugging Milk (and Why Geneticists Are Alarmed)
Instead of long-discounted proxies like skull circumference and family pedigrees, according to experts who track the far-right, today’s proponents of racial hierarchy are making their case by misinterpreting research on the human genome itself. And in debates that have largely been limited to ivory-tower forums, the scientists whose job is to mine humanity’s genetic variations for the collective good are grappling with how to respond.posted by homunculus at 2:45 PM on October 19, 2018
That is so bizarre. More than half the people I know who are lactose-intolerant are white, including me.
posted by bq at 3:15 PM on October 19, 2018
posted by bq at 3:15 PM on October 19, 2018
« Older Grammin' | An Amazing Obituary Newer »
This thread has been archived and is closed to new comments
posted by srboisvert at 2:43 PM on October 16, 2018 [33 favorites]