Genetic testing’s chicken-and-egg problem
September 9, 2008
Speaking as (free) customer of 23 and Me, the genetic testing company co-founded by Anne Wojcicki, the wife of Sergey Brin, the Google co-founder, I can see why it has just slashed its prices.
The New York Times reports this morning that 23 and Me is cutting the price for people who want to spit into a laboratory tube, send it off in the mail, and receive some genetic information about themselves.
Instead of $999 per test, the test price is now $399. Linda Avey, Ms Wojcicki’s co-founder, says that the privately-owned Silicon Valley company wants to entice more people to give it a shot, which would allow more interesting research applications:
“It’s all about numbers and having as many people enrolled as possible.”
Some readers may recall that I was offered a free genetic test while at the World Economic Forum earlier this year, along with my colleague Gideon Rachman. So I have some insight into whether it is worth doing.
My assessment is: at the moment, it is not worth $999. It might be worth, say, $300 if it could get more people to sign up. The problem is that 23 and Me is highly dependent on the network effect: the service improves the more people are covered.
With only my own genetic test to go on, I can learn some mildly interesting things about myself.
I don’t think I am giving away too many secrets by revealing that:
- My mother’s genes come from the H1 haplogroup, a bunch of people who originated 13,000 years ago on the European continent.
- My chromosomes are 99 per cent European and less than one per cent Asian.
- I have typical genetic odds of going bald.
- I metabolise caffeine slowly.
- I am slightly less likely to get bowel cancer than the average European.
As you can see, interesting but not exactly enough to galvanise anyone into rushing out to pay a lot of money to get tested.
Indeed, most people pay for genetic tests to establish whether they are really the father of their children or are related to someone else. This is, in fact, the point. Genetic tests only become interesting when you can compare yourself to others, and preferably lots of others.
I am circumscribed in this respect because the only people I know who have been tested by 23 and Me and agreed to share their results are Esther Dyson, the Silicon Valley guru who is on its board, Gideon Rachman, my fellow columnist, and my wife.
From this, I know that I have no genetic overlap with any of them. It is, incidentally, quite common to have a genetic overlap with your spouse ie to share a relative from thousands of years ago. Ms Wojcicki told me that she has a tiny overlap with Mr Brin.
I also know that my genetic traits, such as susceptibility to various illnesses, are slightly more like Gideon’s than my wife’s. I am not sure what that means.
Overall, however, there are not really enough data to produce any interesting results. If my children, or other relatives, were included then things might be different.
It seems to me that 23 and Me or other genetic testing companies have a problem in this regard. Social networks such as Facebook work because they are free and so you can get your friends to sign up easily. But genetic-based networks such as 23 and Me are held back by the price of admission.
As long as it costs a lot of money to join, few will do it. And, as long as only few people are on the network, its will have limited appeal. To use a genealogical expression, the company has a chicken-and-egg problem.
Back to John Gapper's Business Blog homepage









As GR has not offered to reveal his genetic test result, I am wondering what he has got to hide.
Posted by: J.J. | September 9th, 2008 at 4:09 pm | Report this commentHis genius I suppose?
Dear Mr. Gapper,
From your blog post this week you appear underwhelmed by the results of your 23andme genome scan, which you tried out on a complimentary basis at Davos. You seem unconvinced by what you call the chicken-and-egg logic of their recent blue-light special: If it’s cheap enough, a critical mass of paying customers may sign up and there might in the end be an interesting product to pay for.
To be fair to the Google gang, this may be waiting for eggs to come from a cow. What 23andme is offering is an entertaining social networking service with a double-helix twist. They don’t – and don’t claim to – do any science or even genomic analysis in-house.
But if you think you might be interested in a genome scan that can put you into the context of the genomes of large numbers of others (not just Mr. Rachman, close relatives, odd cousins and Facebook pals), and that can tell you about the very latest correlations to meaningful rather than trivial or amusing traits, I have good news: You don’t have to wait for the masters of the internet to get up to speed on genetics. Unlike the dot-com genetics shops, our service deCODEme (the launch of which last year spurred 23andme to launch hours later) gives you a personal lens through which to look into the biggest and most successful gene discovery undertaking around. You can learn whether you carry risk variants in more than 30 common diseases, discovered and replicated through the analysis of the genomes of hundreds of thousands of people around the world who have taken part in our research over more than a decade. On our blog, www.decodeyou.com, we have stories from some of our subscribers on how they are following up on their deCODEme results with their doctors to improve their health.
If you had been one of our subscribers, just since Davos you would have received updates to your profile on our discovery of novel genetic risk factors for nicotine dependence, lung cancer and prostate cancer. And these updates would not have been made by a consultant downloading our papers from the internet, but from the same scientists that made the discoveries in the first place. Straight – to be true to the barnyard metaphor - from the horse’s mouth.
Edward Farmer
Posted by: Edward Farmer | September 12th, 2008 at 6:50 pm | Report this commentChief Communications Officer
deCODE genetics
There’s no need to pay to be a subscriber to anyone’s service. The Promethease software (www.promethease.com) compares your data - regardless of who produced it - to the SNPedia wiki of known DNA consequences to prepare a personal report for you whenever you want, and it’s all free (for personal use). And SNPedia gains new information daily as research findings are made public.
Of course, another alternative would be to just wait a year or two until you can get your own full genomic sequence at reasonable cost. The DNA chips used by the 23etAl companies survey only a small subset of the genetic variation you actually harbor in your genome …
Posted by: Greg | October 9th, 2008 at 4:43 pm | Report this comment