The Unz Review: An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersGene Expression Blog
Why 23andMe Is No Longer Leading on Personal Genomics
🔊 Listen RSS
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeLOLTroll
These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

23andme_logo I really admire what 23andMe has done. To a great extent they are the “Uber” of DTC personal genomics. FamilyTree DNA really pioneered the sector in the early 2000s, while The Genographic Project scaled things up massively in the middle 2000s. But in the late 2000s 23andMe brought Silicon Valley “disruption” to the game, pushing into disease and traits in a way that both the two earlier efforts consciously avoided. We know how that ended.

But it wasn’t all in vain. 23andMe is today a healthy company, and its shoot-first-ask-questions later actions in the first half of the teens really brought personal genomics into peoples’ lives.

So what’s going with stories like this, 23andMe Has Abandoned The Genetic Testing Tech Its Competition Is Banking On:

For years, genetic-testing startup 23andMe was working to develop a cutting-edge technology that could dramatically expand what its customers might learn about their DNA. While the company’s core product, a $199 “spit kit,” can tell you about your health and ancestry based on small bits of your genetic code, tests based on the new technology — called next-generation sequencing — could provide much more comprehensive information, including your potential risks for many diseases.

But 23andMe has given up on the technology for now, BuzzFeed News has learned.

I think one way to understand what’s going on is that though the firm’s consumer face is still as a DTC personal genomics outfit, it is really banking on becoming a genetically savvy pharmaceutical corporation. Genomics is the future, but pharm is the present.

51cz5E2hKTL._SX378_BO1,204,203,200_ 23andMe probably has ~1.5 million genotypes now. They’ll confirm more than 1 million. If they had more than 2 million I assume they would tell us they did. What are they doing with those genotypes? It was always understood by most that 23andMe was increasing its database to the point where they could generate associations that academics could not because of lack of statistical power. The problem now, with more than 1 million genotypes, is that they need phenotypes.

It is much more valuable for 23andMe to get rich data on one customer, than it is to gain one hundred more random genotypes. That’s probably why they’re not sweating that the $199 price point discourages people, especially when those people are getting less than they did in the past. That’s also why they are pulling out of the game in next generation sequencing. Sequencing is basically a commodity business now, and just not as good a return on investment as gearing toward the pharmaceutical market. Sequencing deeply has some benefits, but there is no way 23andMe would be able to subsidize the $1,000 cost of a good 30x genome to get enough of a sample size to return the investment.

None of this is a big secret. A friend of mine was talking about this in the broadest sketches at the 23andMe party at ASHG.

 
• Category: Science • Tags: 23andMe, Genomics 
Hide 7 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. I’m trying to understand – Do you think their business practices were OK, because the ends justify the means?

    Several years ago, I was perplexed by people such as you & D. MacArthur seeming to be on friendly terms with 23andme, when you probably would have considered their behavior reprehensible coming from a bank, big pharma, or other corporation.

    • Replies: @Razib Khan
    a bank, big pharma, or other corporation.

    ineffective medicine can kill people. banks can bankrupt people. 23andMe dealt in information, often quite innocuous. the diet and supplement industry cause much more harm than 23andMe and other genomics firms ever could.
  2. @O'really
    I'm trying to understand - Do you think their business practices were OK, because the ends justify the means?

    Several years ago, I was perplexed by people such as you & D. MacArthur seeming to be on friendly terms with 23andme, when you probably would have considered their behavior reprehensible coming from a bank, big pharma, or other corporation.

    a bank, big pharma, or other corporation.

    ineffective medicine can kill people. banks can bankrupt people. 23andMe dealt in information, often quite innocuous. the diet and supplement industry cause much more harm than 23andMe and other genomics firms ever could.

  3. What I do not understand: does 23andMe use effective persuasion techniques to get its customers to give it health data? Sure they have SNP info on one 1 million customers. But how do they get enough other data to correlate with it? Or does just having 1 million sets of SNPs give them something useful for pharma?

  4. It doesn’t seem to me that their number of genotypes has doubled in the last 5 years, because I have not seen a doubling of the number of ‘3rd cousin’ relationships in the family finder (other than family that I submitted myself.) And I manage accounts for about 20 unrelated people as well.

    As for high quality sequencing and phenotypes, the smartest thing to do would be to specifically subsidize full sequencing for families where many individuals agree to give lots of phenotype data.

    I could easily convince my parents, my siblings, and my aunts and uncles to provide any kind of phenotype data asked of them. Probably even medical records. And I would do it if we were given the full data set at the end.

    I think this would be true for several thousand families at least, and especially families with lots of health problems, and families without any health problems.

    • Replies: @Razib Khan
    unless they are lying, it's gone up by an order of magnitude. probably they tweaked the algorithm or the engineering to cap how many matches you get.
  5. @Rick
    It doesn't seem to me that their number of genotypes has doubled in the last 5 years, because I have not seen a doubling of the number of '3rd cousin' relationships in the family finder (other than family that I submitted myself.) And I manage accounts for about 20 unrelated people as well.

    As for high quality sequencing and phenotypes, the smartest thing to do would be to specifically subsidize full sequencing for families where many individuals agree to give lots of phenotype data.

    I could easily convince my parents, my siblings, and my aunts and uncles to provide any kind of phenotype data asked of them. Probably even medical records. And I would do it if we were given the full data set at the end.

    I think this would be true for several thousand families at least, and especially families with lots of health problems, and families without any health problems.

    unless they are lying, it’s gone up by an order of magnitude. probably they tweaked the algorithm or the engineering to cap how many matches you get.

    • Replies: @Rick
    That makes sense. I see that they give much wider predictions of relationship now than in the past.

    However, I still have people on 23andme (white Americans with all of their ancestry from places like The Netherlands and Northern Ireland) with fewer than 5 relative matches sharing more than 0.4% total on only 1 matching segment. This is what they now classify as 3rd-6th cousins.

    I find it quite remarkable that people of European descent could have so few 'close' relatives from a database of 2 million people, most of which have the same background.
  6. @Razib Khan
    unless they are lying, it's gone up by an order of magnitude. probably they tweaked the algorithm or the engineering to cap how many matches you get.

    That makes sense. I see that they give much wider predictions of relationship now than in the past.

    However, I still have people on 23andme (white Americans with all of their ancestry from places like The Netherlands and Northern Ireland) with fewer than 5 relative matches sharing more than 0.4% total on only 1 matching segment. This is what they now classify as 3rd-6th cousins.

    I find it quite remarkable that people of European descent could have so few ‘close’ relatives from a database of 2 million people, most of which have the same background.

  7. If 23andMe is no longer the best value in consumer testing, which company is?

Comments are closed.

Subscribe to All Razib Khan Comments via RSS