5 ms·

What you're claiming about the privacy of DNA isn't strictly true. You're correct that, aside from mutations, we share all of our alleles with one or the other

by rtl49 11y ago

What you're claiming about the privacy of DNA isn't strictly true. You're correct that, aside from mutations, we share all of our alleles with one or the other parent, and about half of them with full siblings, etc. But this information in itself doesn't enable one to predict the exact composition of your genome, which would be needed to form a predictive model of your body. It would still be possible to, say, link your DNA to that found at a crime scene, but you still sacrifice something by giving it up. I have little hope that we will maintain this form of privacy, but that isn't to say nothing has been lost.

eggie 11y ago
We have not had this form of privacy for a long time, and I doubt we will regain it unless we would like to do away with DNA evidence altogether. If anything we should be pushing for more precision in the use of DNA samples as evidence. The standards required for identification are incredibly low and result in large numbers of false positives. These poor standards cause harm. The correct use of DNA evidence can clarify crimescene events and has had the power to exonerate many falsely-accused people.
- rtl49 11y ago
  I think these are separate issues. There is definitely an important difference between being able to use DNA in criminal investigations and maintaining a database of every genome in the population. The first solves a problem of legitimate social concern, the second is the stuff of dystopian fiction. None of this is at odds with insisting upon higher levels of precision and integrity in forensic science. We can talk about reducing the number of false positives and educating juries about the limitations of forensics without resigning ourselves to the unwelcome collection of our DNA.
  - JoeAltmaier 11y ago
    Yet DNA is so easy to collect; you don't need permission. I think we'll have to get used to the idea of universal identification. Its inevitable. Better to come to terms with it.
    - rtl49 11y ago
      I disagree that it's better to come to terms with it. If we were to normalize a sense of disgust at the practice we could institute legal protections against bulk collection. What you're describing is simply a world that few people want.
      - JoeAltmaier 11y ago
        Not wanting it, and able to prevent it, are different. I maintain we cannot prevent it. All we have left is coming to terms.
        rtl49 11y ago
        Like I said before, "I have little hope that we will be able to maintain this form of privacy." My dispute is that it isn't worth trying to save, or that it's a thing of little value.
eggie 11y ago
Another point, which is that although the precise composition of alleles in one individual is not known, even a handful of markers are enough to identify a member of a given family. People will share their genomes because sharing is incredibly beneficial and the risks are indistinguishable from 0. As long as people care about genealogy it will be difficult to prevent identification of your genome at a family level. This has been demonstrated for a set of anonymized genomes that were part of public genomics projects. http://www.sciencemag.org/content/339/6117/321.short http://www.sciencemag.org/content/339/6117/321.short > Sharing sequencing data sets without identifiers has become a common practice in genomics. Here, we report that surnames can be recovered from personal genomes by profiling short tandem repeats on the Y chromosome (Y-STRs) and querying recreational genetic genealogy databases. We show that a combination of a surname with other types of metadata, such as age and state, can be used to triangulate the identity of the target. A key feature of this technique is that it entirely relies on free, publicly accessible Internet resources. Unless we make sharing genomes illegal (and we absolutely should not!) this is going to become a hard fact of life. So we should focus on mitigating potential problems that arise in different ways.
- rtl49 11y ago
  I wasn't referring to identification (I even specifically mentioned that with respect to forensics). I'm talking about knowing the specific content of a subject's genome, the issue of primary concern. Even if you know the genomes of both parents, you can only evaluate the probability of their offspring possessing any particular combination of alleles. The more alleles you wish to know about, the lower the probability that any particular model is correct. Regardless, you seem to completely disregard the argument for legal protections. If there were laws guaranteeing an absolute right to control one's genetic information, including who may possess it, much of this concern would go away.

by rtl49 11y ago

eggie 11y ago

We have not had this form of privacy for a long time, and I doubt we will regain it unless we would like to do away with DNA evidence altogether. If anything we should be pushing for more precision in the use of DNA samples as evidence. The standards required for identification are incredibly low and result in large numbers of false positives. These poor standards cause harm. The correct use of DNA evidence can clarify crimescene events and has had the power to exonerate many falsely-accused people.

rtl49 11y ago
I think these are separate issues. There is definitely an important difference between being able to use DNA in criminal investigations and maintaining a database of every genome in the population. The first solves a problem of legitimate social concern, the second is the stuff of dystopian fiction. None of this is at odds with insisting upon higher levels of precision and integrity in forensic science. We can talk about reducing the number of false positives and educating juries about the limitations of forensics without resigning ourselves to the unwelcome collection of our DNA.
- JoeAltmaier 11y ago
  Yet DNA is so easy to collect; you don't need permission. I think we'll have to get used to the idea of universal identification. Its inevitable. Better to come to terms with it.
  - rtl49 11y ago
    I disagree that it's better to come to terms with it. If we were to normalize a sense of disgust at the practice we could institute legal protections against bulk collection. What you're describing is simply a world that few people want.
    - JoeAltmaier 11y ago
      Not wanting it, and able to prevent it, are different. I maintain we cannot prevent it. All we have left is coming to terms.
      - rtl49 11y ago
        Like I said before, "I have little hope that we will be able to maintain this form of privacy." My dispute is that it isn't worth trying to save, or that it's a thing of little value.

eggie 11y ago

Another point, which is that although the precise composition of alleles in one individual is not known, even a handful of markers are enough to identify a member of a given family. People will share their genomes because sharing is incredibly beneficial and the risks are indistinguishable from 0. As long as people care about genealogy it will be difficult to prevent identification of your genome at a family level. This has been demonstrated for a set of anonymized genomes that were part of public genomics projects. http://www.sciencemag.org/content/339/6117/321.short http://www.sciencemag.org/content/339/6117/321.short > Sharing sequencing data sets without identifiers has become a common practice in genomics. Here, we report that surnames can be recovered from personal genomes by profiling short tandem repeats on the Y chromosome (Y-STRs) and querying recreational genetic genealogy databases. We show that a combination of a surname with other types of metadata, such as age and state, can be used to triangulate the identity of the target. A key feature of this technique is that it entirely relies on free, publicly accessible Internet resources. Unless we make sharing genomes illegal (and we absolutely should not!) this is going to become a hard fact of life. So we should focus on mitigating potential problems that arise in different ways.

rtl49 11y ago
I wasn't referring to identification (I even specifically mentioned that with respect to forensics). I'm talking about knowing the specific content of a subject's genome, the issue of primary concern. Even if you know the genomes of both parents, you can only evaluate the probability of their offspring possessing any particular combination of alleles. The more alleles you wish to know about, the lower the probability that any particular model is correct. Regardless, you seem to completely disregard the argument for legal protections. If there were laws guaranteeing an absolute right to control one's genetic information, including who may possess it, much of this concern would go away.