Comment by karaterobot | Hacker News Reader

Comment by karaterobot | Hacker News Reader

https://archive.ph/CeLJZ

After reading the article, it looks to me like giving access to this data is not what's new, it's that the contractual terms of the existing agreement have changed. That is, it seems like 23AndMe already gives access of some form to pharmaceutical research companies.

Two questions I had:

* I'm vaguely familiar with the challenges of anonymizing omics data, and I'm wondering how they expect to make someone's entire genetic profile sufficiently anonymous or deidentified. All I've seen from them is one of their example reports, and you could take the name off one of those, and it seems like it would still be pretty easy to reidentify the subject just from that high level report, let alone from the raw data.

* When they say these are users who have opted in to data sharing, I wonder what form that actually takes. Is this one of those things where everybody who uses 23AndMe is opting in? Did these 14 million users all actually provide informed, active consent? I assume that to maximize the value of that data, 23AndMe is incentivized to interpret consent as broadly (and vaguely) as possible.

My biggest concern is that my siblings and parents have all signed up. So (more or less) my DNA is in this database, even though I have never interacted with the organization.

Yes, I wrote about that when this all first started happening. Not that anybody cared... it doesn't take a whole lot to be able to identify you given some samples of your extended family. And as a means of elimination it will be extremely effective.

Gattaca is an amazing movie, it has aged so well I find myself having more and more appreciation for the amount of thought that went into the script.

Gattaca is about the senselessness of the dystopia, not the inevitability of it!

Yes, I know what it is about. But what gets me is how incredibly prescient it all is, we're pretty much running headlong into this future without any ethical framework to guide any of it. It's a free-for-all right now, as if the new frontier of the Wild West re-opened. Reminiscent of the early days of the web. But Gattaca seems to be used as a blueprint rather than a warning.

The horrors of using information for research?

I'm on board with the idea that it's going to be difficult to make the opt in on something like this respectful of the actual wishes of the participants, but they are at least partly working in that direction.

No, the horrors of the potential downsides associated with the abuse of that information and the fact that once it is out of the bag you can't put it back in.

It's never about the data itself, it's about what the worst possible groups can do with it given some nation state backing.

> But Gattaca seems to be used as a blueprint rather than a warning.

There's a founder of a genetic sequencing and embryo selection start up that I talked to who said they were "inspired by Gattaca" and isn't shy about bringing that inspiration up[1] at any opportunity.

Their company's promotional material also isn't shy about selecting embryos for predicted intelligence, height, etc.

[1] https://nypost.com/2019/11/09/genetic-test-aims-to-predict-a...

We keep creating the Torment Nexus. Because someone can profit from it, therefore someone must profit from it.

> Did these 14 million users all actually provide informed, active consent?

Virtually impossible, even without preconceived notions of this company's ethics (which... should be obvious), because it's actually very difficult to get informed consent. Not only does it take longer and more work to get the consent, you'd actually have to work pretty hard to define a system that >90% of people would consider true informed consent. There's no blueprint for it in the corporate world. The default consent-getting is so bad that even doing a bad job can look amazingly progressive in comparison.

23&me's consent form is excellent and a paragon for the field. https://www.23andme.com/about/consent/

It's not clear to me this takes automatic effect, and actually states "If you agree to this consent" - I assume that users are given the option to consent or not?

This seems to be the usual TOS - https://www.23andme.com/privacy/ and it says "We will not share your genetic data ... without your explicit consent."

So the news here is "Drugmakers are set to pay 23andMe to access consumer DNA for consumers that have given their explicit consent to do so". Maybe the fact 23am is getting paid is considered a sour note here, but it doesn't seem this is otherwise underhanded.

No, there's nothing underhanded here. 23&Me has never hidden its long term intention to use genetic data, from informed consent customers, to help solve diseases. To me the idea was obvious from before their founding, and it's amazing how long it took them to get to this point.

Reply on news.ycombinator.com