Genomics and me: DNA testing may not reveal as much about yourself as you'd hope | Hacker News Reader

Genomics and me: DNA testing may not reveal as much about yourself as you'd hope

thelongandshort.org

30 pointsjeremynicolas23 comments

> Two years ago, US consumers could see more than 240 detailed health and trait reports, including risk factors for heart attack and type 2 diabetes. Then, at the end of 2013, the Food and Drug Administration clamped down. They implied that 23andMe's tests were inaccurate. They also worried that some consumers would misconstrue the risks, seeking out unnecessary treatments and changing their medication regimes without speaking to their doctors.

I can't even begin to describe how much this enrages me: I, a purportedly-free citizen of a purportedly-free republic, am not permitted to pay a firm to do its best to give me accurate information, because some unelected third party is concerned that I may make bad decisions based on that information.

What business of the State's is it if someone chooses to change his medical regimen without consulting his physician? It's his regiment, not the State's; it's his health, not the State's.

I'm a free man: I should be permitted to consult the best firms and practitioners I can in order to make decisions for my own lifestyle.

The business of the state is to prevent false advertising, because it deprives the citizens of making informed decisions, and because it awards financial resources to those who are willing to profit by harming others.

It happens so that for health-related issues the bar for false advertising is higher than the for other products. I think it's for a good reason. The authority to set the bar was given to the FDA by the people's elected representatives.

The FDA started out with the best of intentions--to stop snake oil peddlers or food vendors who sold fatal concoctions as "delicious" or "miracle cures."

23andMe is not selling any drug. The FDA is now in the business of regulating information that goes to the consumer. And the FDA has increasing power with no oversight. Potentially life-saving medicines have to jump through decades of clinical trials before they can be used to save lives.

I'm sure you appreciate the other side of that argument – that organisations should not be able to advertise medical devices and services without demonstrating that the results are accurate.

The idea that a medical services company and an individual consumer are on the same footing doesn't carry much weight for me. One of the obvious solutions is for a state body to set a standard for what accuracy is required in order to make medical services available; this is a pretty ideologically coherent attempt to address the fundamental inequality there.

I'm happy to hear an argument that the FDA over-regulates, or that other regulatory approaches could be more beneficial. But a complete free-for-all invites a return to the days of snake-oil patent medicines – and I don't think that's something anybody wants to see.

The FDA implied that the tests are inaccurate. Where's their evidence?

They have no oversight--they are the oversight. That is a problem. They can claim anything, and use it as an excuse to add arbitrary regulations and rules that have an end result of pleasing lobbyists or voters or the latest celebrity-politician.

I think it's a complete tragedy that customers in the US can no longer get health information from 23andMe.

I started a genome software company that primarily targets the academic research market, but we recently adapted our software for 23andMe users: https://www.enlis.com/personal_edition.html

Happy to hear opinions and bug reports!

How does the report compare to the one generated by Promethease[1]? Promethease itself uses data from SNPedia[2]

[1] https://promethease.com/

[2] http://www.snpedia.com/index.php/Promethease

Any idea where I can get a full sequencing at a reasonable price for personal use?

While I somewhat agree people should be able to have their DNA tested freely if they so choose, perhaps these regulations are wiser than we are aware at the moment. Look at what people foolishly give Facebook. Will these same people be more careful with their DNA information? Will this information be used against them at some point in the future? Just something to consider.

Anybody who finds a flake of your skin can have your DNA pattern; its not going to be something that can be kept secret. We'll have to learn to live in this brave new world, with such wonders in it.

#myDNAmyData

This brings up a number of problems with the current approaches to gathering signal from the noise of DNA, but it's making the deeper argument that DNA may not contain much useful information. I don't think it met the burden of proof for that argument. It's a popular argument: we want to believe that our attributes aren't predetermined, that we have free will, so we want to believe anything that downplays the role of nature in the nature vs. nurture debate. But really this is still a big unknown. I very much doubt that the difficulties we currently have with gathering useful information from DNA are unsolvable.

I think one of the problems is that the sequencing boon has led to a shift in grant funding and research direction towards global hypothesis generating studies rather than hypothesis driven studies. These may find statistical associations between DNA variations and phenotypes but tend to avoid doing the difficult detailed characterizations that will help us understand the functional / molecular context.

True. You do have to generate hypotheses before you can use them to drive studies, though. Those statistical associations enable the detailed studies by telling them where to look.

> we want to believe that our attributes aren't predetermined, that we have free will, so we want to believe anything that downplays the role of nature in the nature vs. nurture debate

Who wants to believe this? It seems pretty lop-sided to accuse the "nature"-skeptics (or "nurture"-proponents) of falling prey to confirmation bias, when our history is littered with much more evidence of the atrocities committed by eugenicists.

Even today, it seems that many people who are successful are desperate to find out how much it is their innate destiny, due to genetics. It certainly provides great justification for providing solely for yourself and your offspring instead of for caring for a big social safety net and whatnot.

> Who wants to believe this? It seems pretty lop-sided to accuse the "nature"-skeptics (or "nurture"-proponents) of falling prey to confirmation bias, when our history is littered with much more evidence of the atrocities committed by eugenicists.

Ugh, eugenicist atrocities have literally nothing to do with the science. Believing that nature is the leading cause of characteristics does not justify eugenics. And even if this did, how on earth does this disprove that nurture proponents are experiencing confirmation bias?

I'm not a nature proponent (or a nurture proponent). The fact that many nurture proponents experience confirmation doesn't prove them wrong; I'm merely saying that they're believing what they believe for the wrong reason.

> Even today, it seems that many people who are successful are desperate to find out how much it is their innate destiny, due to genetics.

This does not in any way disprove that some nurture proponents are experiencing confirmation bias.

Overall you've made this gigantic logical leap that because a small subset of nature proponents did bad things, nurture proponents must not be experiencing confirmation bias.

And before you make yet another logical leap and assume I'm talking about all or most nurture proponents experiencing confirmation bias, I'm not. I've said "we" to talk about a common human motivation, but I'm really talking about what I think the motivation of the author of the article is. I'm not making any claims about other people.

DNA is just a small part of what makes up the functioning whole of an organism. The Transcriptome, Proteome, and the Metabolome are just as important to determining the outcomes of genes.

It's like understanding what parts are in your car but not knowing how much gas is in the tank, how far the pedal is pressed, or even what gear you're in. It's still an incomplete picture.

Disclosure: I studied cancer metabolism in grad school so I'm a little biased to think that the Metabolome is the functional endpoint of genetics and all its associated modifiers.

Well, in the nature vs. nurture debate I'm not really interested in taking a side; I want to figure out the truth. I'm relatively certain that it's a split. I just want to caution against a common human bias in favor of nurture, which this article seems to be affected by.

To anybody who has 23&Me results: can you tell me if they are still reporting tongue rolling as a heritable trait on their reports?

I had it done around 2013 or 2014, and yes, that was in the report.

Sorry to pry, but is it possible you could go back and look at the report and give an indication of how strongly it seems to imply the trait is genetic? Does it include any caveats (without following their links to literature)? Otherwise indicate this trait isn't strongly genetically determined?

Presumably you are aware this is a Myth of Human Genetics?

http://snpedia.com/index.php/Myths_of_human_genetics

Yes, I've read that paper, and I read 23&Me's blog post on the trait from 2008 (http://blog.23andme.com/23andme-and-you/tackling-tongue-curl...). They both agree in saying that tongue rolling is not a Mendelian trait, and they both agree there appears to be a heritable component which affects several tongue-shaping traits.

I'm asking because the 23&Me reports don't really include the contextual information to appreciate the subtlety of non-Mendelian traits. If people see this and come to the false conclusion, then they're likely to come to similar false conclusions abotu the parts of the report that touch on actual health-related issues.

Reply on news.ycombinator.com