Forgot your password?
typodupeerror
Medicine Science

Renowned Geneticist Analyzes Consumer DNA Tests 97

Posted by kdawson
from the all-in-the-interpretation dept.
pdragon04 sends in the hardly surprising news that direct-to-consumer genetic testing isn't predicting diseases as well as they claim. "...[Francis] Collins, who played a central role in the Human Genome Project and is rumored to be the next head of the National Institutes of Health, announced at the Consumer Genetics Conference in Boston last week that he had had his genome analyzed [using a made-up name] by the big three of direct-to-consumer genetic testing: 23andMe, Navigenics, and DecodeMe. Collins said that sequence-wise, the tests 'appear to be highly accurate': there were almost no differences in the genotype information generated in the three different analyses. But there were significant differences in the numbers of genetic variations used to calculate disease risk, as well as the final risk score. ... For example, one company used 5 single nucleotide polymorphisms, or SNPs, to calculate risk for a particular disease, pronouncing Collins at low risk. Another used 10 SNPs, placing him at high risk, and the third used 15, concluding that he is at average risk."
This discussion has been archived. No new comments can be posted.

Renowned Geneticist Analyzes Consumer DNA Tests

Comments Filter:
  • sequence once (Score:5, Informative)

    by jmilloy (1497261) on Tuesday June 16, 2009 @10:54PM (#28357483)
    Sounds like there's room in the market for just the risk analysis. No reason to have the dirty work done three times - just sequence once and get a whole range of opinions, or focus on certain areas for detailed analysis. maybe this already exists.
    • Re:sequence once (Score:5, Insightful)

      by jd (1658) <{moc.oohay} {ta} {kapimi}> on Tuesday June 16, 2009 @11:45PM (#28357723) Homepage Journal

      To judge from the complaints that frequently get posted to the GENEALOGY-DNA mailing list, even the experienced big-name DNA corps screw up on even testing a few dozen STRs. The chances of fouling up somewhere along the entire genome seems extremely high. Thus, redundant testing seems a really good idea - at least until they improve their methods.

      On the other hand, there probably IS a huge market for risk analysis. The DNA analysis market is pretty much glutted, but there's not much out there for DNA analysis for the average person at the moment.

      It is worth bearing in mind that the sorts of studies done might not pick up all of the secondary genetic data needed - markers whose presence doesn't alter the probability of a condition UNLESS the primary marker is also present. Simply doing a gigantic "diff" isn't going to pick those up, as their presence isn't always important.

      Then, there are conditions with multiple genetic triggers. Chronic fatigue has seven, according to a BBC report a while back, each of which produces a different condition with essentially identical symptoms but which respond to different treatments. (Unless there's genetic overlap, I would assume this means there are 5040 potential conditions that could be produced from these markers.)

      Then, on top of all that, some conditions may occur but be too mild to notice. There are claims that autism is far more common than currently thought and certainly, the UK diagnoses it twice as often as the US (implying that either closer examination DOES yield more positives, or that us British are just strange). There's also a claim that many, if not most, people have some degree of synaesthesia. But is it useful to know that there's a very high risk of being normal?

      (Ok, ok, Slashdotters might want to know if there's a genetic danger of becoming normal, but ASIDE from that...)

      It would seem to me that conditions such as autism and synaesthesia follow something close to a Poisson distribution, with the vast bulk of people at close to no effect. Even if they're absolutely guaranteed the effect, they'll never notice.

      Unless studies into risk go far, far beyond the horribly basic and naive results you get at the moment, the results are useless. And that kind of massive data-crunching analysis is something that Beowulf clusters and BOINC clients are going to be very good at... ...provided an answer even exists. If the severity isn't present in the DNA, then the risk is useless.

  • If I'm going to drop a couple Gs on getting a genome-based risk assessment, they had damn well better deep-sequence that shit.
  • So which service produces, in his expert opinion, the better overall results/conclusions/advice?

    I RTFA and fail to see what the pros/cons of each service is, and which one he recommends.

    I know that would be a boon to whichever company is producing the better results, but who cares? If he is suppsoed to be independent, what does it matter if he gives his honest opinion?

    • I RTFA and fail to see what the pros/cons of each service is, and which one he recommends.

      That's because he can't recommend any on the basis of the results or methodology of the testing.

      SNPs can be a useful tool, but because (by definition) they are single nucleotides, you need a lot of them to be able to validate a pattern. From the submission, he mentions 3 labs using 5, 10 and 15 SNPs and claiming to be able to draw inferrences from these. My understanding (FWIW) is that a mere 5 polymorphisms are not
      • Problem is, he doesn't specify which labs use more SNPs or "more useful" SNPs (which as you mentioned, should provide better/more accurate results).

        The point of experiments and tests should be to draw some conclusion (even if the conclusion is that all fail) but clearly some services are performing better than others and are turning out more accurate. Why does he not state, in his professional and unbiased opinion, which one is better?

        • Problem is, he doesn't specify which labs use more SNPs or "more useful" SNPs (which as you mentioned, should provide better/more accurate results).

          Primarily because it's not known yet for a large number of diseases.

          The point of experiments and tests should be to draw some conclusion (even if the conclusion is that all fail) but clearly some services are performing better than others and are turning out more accurate. Why does he not state, in his professional and unbiased opinion, which one is better?

          Becau

  • ...and let geneticists sort us out.

  • by Anonymous Coward

    sequence-wise, the tests 'appear to be highly accurate': there were almost no differences in the genotype information

    Almost no differences? Can't a difference in a single gene do pretty crazy things? Like your-gene-is-alenine-don't-worry-you'll-be-just-fine. That gene-is-a-glutamate-your-eyes-will-number-eight...

  • Clearly the statistical analyses' are wrong for at least two of those companies. The prior probability of risk for a given disease is 'average', and if you don't test enough polymorphisms or if the correlations are weak then it remains average. Trouble is you can't make a business case on selling such weak information, so there's an incentive to spice up the summary info they provide.

  • by MobyDisk (75490) on Wednesday June 17, 2009 @07:10AM (#28359981) Homepage
    The practice of using different numbers of gene sequences is common: the same thing happens if you get an HIV test, or an HPV test, or FLU, or whatever. In that case though, the FDA regulates it to prove that the result is clinically valid. I'm not sure what involvement the FDA has in this.
  • Heritability (Score:2, Insightful)

    One item in the article that surprised me: the companies aren't offering information to their clients about diseases they are carriers for. For instance, it would add value to their service if clients knew they carried the gene for cystic fibrosis (a common genetic test).

    It's either a huge oversight by the 'big three', or they think that their clients are so focused on themselves as to not really care about what diseases their children could inherit.

    • by FleaPlus (6935)

      One item in the article that surprised me: the companies aren't offering information to their clients about diseases they are carriers for. For instance, it would add value to their service if clients knew they carried the gene for cystic fibrosis (a common genetic test).

      That's kind of odd, because at least 23andme (haven't checked the others) seems to offer information on whether someone is a cystic fibrosis carrier:

      https://www.23andme.com/health/cysticfibrosis/ [23andme.com]

      Cystic Fibrosis (Delta F508 mutation)

      Cystic Fibrosis is one of the conditions that 23andMe analyzes. Our service includes the following information:

      * Whether or not you are a carrier for the Delta F508 mutation linked to cystic fibrosis.
      * Information on i3000001, a marker that influences your carrier status for Cystic Fibrosis.

      * Background information on Cystic Fibrosis and a list of counselors, links and support groups in your area.

  • DNA Sequencing Report for: John Doe Sequencing: 100% Complete Risk for all diseases: average. Have a good day. :)
  • I began performing SNP analysis on genomic DNA for schizophrenic patients back in 2002 in the new Pharmacogenoics group of a large Pharmaceutical company. Our goal was to correlate drug effect to the genetic background of the individual. If we could establish a relationship we could potentially get a few failed drugs to perform "better" when used in patients genetically predisposed for a positive effect and get them into the market.

    All of our results required TEAMS of statisticians to figure out if the
  • The defense lawyers used widely varying statistical numbers from scientific experts to create doubt as to whether OJ Simpson's genes matched blood found at the crime scene. This was one piece of doubt among several others in the case before a jury eager to find any reasonable doubt to acquit.

Your own mileage may vary.

Working...