[m12k]

I'm really curious about what I could learn by getting my DNA sequenced, but I'm worried about my rights to not have it recorded and shared without my consent if I got someone else to do it for me - so any advance toward an affordable home test setup is very welcome.

[adabaed]

Imagine insurers refusing to give you a service due to your predisposition to certain diseases...

[meltedcapacitor]

Protection from this comes from laws that ban DNA-based policies, not by being secretive about sequencing. If it is allowed, insurers will have no need to obtain DNA sequences in devious ways, they will just ask and refuse cover or charge more when clients refuse to get sampled.

[m12k]

Sure, but being secretive about your DNA seems like the prudent course of action until those laws are in place

[toomuchtodo]

“Passed in 2008, a federal law called the Genetic Information Nondiscrimination Act (GINA) made it illegal for health insurance providers in the United States to use genetic information in decisions about a person's health insurance eligibility or coverage.”

Also prevents employment discrimination based on genetics.

https://www.genome.gov/about-genomics/policy-issues/Genetic-...

(disclosure: have had my DNA sequenced by multiple organizations, and it's publicly available)

[jrumbut]

What I worry about is having this data laundered through a couple of vendors.

"How could we know our vendor's vendor was using genetic information in their proprietary risk score?"

"How could we know our client's client was using our score for life, health, or auto insurance/employment/lending/etc decisions?"

It's a "can't unring a bell" situation and the gaps in the regulations and the incentives for bad behavior are enormous.

[andrewxdiamond]

Because insurance is regulated against this. They can’t just Willy-nilly get data and “scores” from uncontrolled sources.

[sathomasga]

Note that the law does *not* apply to other types of insurance. Life insurance, for example

[ajuc]

It's amazing how many problems you avoid by having public health system.

[adabaed]

You resolve part of them, but immediately generate others. Hybrid systems are the way to go.

In Spain, for example, we have a private system but it is extremely inefficient in some areas (and very good in others). Of course, you can have private insurance, but you still have to pay your social security. Curiously, the only ones who can decide which system they want are the public servants...

[siegfried-en]

It's exactly the same in Portugal.

[inglor_cz]

You avoid the problem with medical debt, to be precise.

You cannot really avoid the fundamental constraints - anywhere in the world, there are only so many doctors and so much money available for treatments. IDK if USA has a shortage of doctors, but plenty of European countries do. A country like Romania just cannot give its doctors big enough wages to stop them from seeking employment elsewhere, where they will get five to ten times as much (UK, Germany, Switzerland). As a result, local hospitals are seriously understaffed.

Where I live, having personal connections to good doctors gives you an advantage - you will be examined and treated faster. Then there is outright nepotism.

The outgroups are different than in America, but there are always people for whom the system sucks.

[mindslight]

That's a pretty poor way of pigeonholing the problem. Looking at the US healthcare system, it's obvious that many doctors' and nurses' talents are wasted doing bureaucratic paperwork. Simultaneously, if there is a genuine lack of healthcare providers, there is no price signal that would encourage more to enter the market.

What you say may be somewhat true in the context of transmuting the US's "private" bureaucracy into bona fide "government". But it's certainly not a "fundamental constraint" that's impossible to solve. Rather it's a failure of organization, whether critiqued in terms of bottom-up market failure or top-down governance failure.

[kortilla]

> Looking at the US healthcare system, it's obvious that many doctors' and nurses' talents are wasted doing bureaucratic paperwork.

This is incorrect. Most of the paperwork is done by administrative staff. Paying for that giant staff + the actual medical professionals is why things are so expensive.

Hospitals are not stupid, they won’t waste their most valuable resource (healthcare time) on bureaucratic paperwork.

[mindslight]

An oncology department I'm familiar with has an entire "nurse navigator" whose whole job is to submit "prior approval" requests to "insurance" companies justifying why patients need a specific treatment, plus the nurses employed by the "insurance" companies reading those requests. I believe it's similar for any moderately expensive specialty. A common career path is care nurse -> burnout -> administration. Most of the administration is made up of people who could be providing healthcare.

And no, hospitals' most valuable resource are their billing computers. I think when it comes to providing actual healthcare hospitals are very stupid. You cannot partition any knowledge worker's attention into 10 minute blocks and expect them to achieve anything useful, yet that is what their entire system is designed around. The hospital doesn't have unilateral say of course (an "insurance" company won't pay one doctor the "price" of two if they spend twice as long with a patient), but they're still content optimizing within that status quo outcome - completely scatterbrained care.

And it's not like individual doctors are well rested or happy when you talk to them. The system clearly takes their toll on them (eg disappearing for 5 minutes to go retrieve test results that didn't show up before your appointment). In fact I'd say the vast majority of human talent in the medical system ends up completely wasted.

[candiodari]

Capitalism, nepotism, public, private, insurance, nationalised healthcare ... the GP is saying that these are methods of dividing up available care, not methods for creating more available care.

X is the amount of medical care available

Y is the amount of medical care wanted

If Y < X there is no problem with any of the systems. And, obviously, a certain amount of inefficiency doesn't affect patient care. Plus, perhaps relevant today: when shit hits the fan we can scale up available care quickly.

If X > Y it doesn't matter which system you choose, someone will go without. You can change who goes without but you cannot fix the system by changing the method of dividing care.

Can things be somewhat improved with better organisation? Sure. Probably. But let's not overestimate it either. Let's take a dream scenario: optimal organisation can make 20% more care available. How much more care is wanted? I think we can safely say the US population wants 200% or more than the current system provides. Whilst nobody's opposed to improving organisations, it cannot fix the problem.

Fixing the problem is something you can only do by doubling the medical training available. That'll be a lot of extra dollars, none of which go anywhere near patient care for at least 10 years, so I would expect a lot of strong opposition from a lot of sides. But it's the only way to fix things.

[ajuc]

Looking at spending per capita it's clear that American problems aren't caused by lack of money in the system.

They are caused by high barriers of entry, which in turn are caused by entrenched elites gatekeeping jobs through absurdly high tuition fees, expecting everybody to take lots of student debt and a very litigation-friendly environment. These costs are then passed on to the general population through a byzantine system of health insurance that leaves a lot of people uninsured.

> Let's take a dream scenario: optimal organisation can make 20% more care available.

In 2020 UK spent 3278 GBP (~4400 USD) per capita on healthcare [1]. USA: 12,530 USD. That's about 3 times less or a difference of 200% [2].

In UK life expectancy is 81.2 years. In USA it is 78.79 years.

3 times more spent to get a worse outcome doesn't seem like "20% difference" to me. Of course there are other factors, but are they enough to overcome 3x difference? I don't think so.

You cannot compare healthcare systems on X doctors per Y patients basis, because the outcomes aren't linear. It's orders of magnitude more expansive to treat many health problems if you go to the doctor 2 years too late. And the outcomes are worse despite the higher costs. Guess what happens when people have to pay a lot for each visit - often they go too late.

[1] https://www.statista.com/statistics/472940/public-health-spe.... [2] https://www.cms.gov/Research-Statistics-Data-and-Systems/Sta...

[monopoledance]

The argument was about "free-market" insurance contracts tho.

[foobarbecue]

If you haven't seen Gattaca, you should

[haihaibye]

There should be a directors cut where the mission fails because of Vincent's hidden heart condition.

Gattaca shows eugenics has been so vilified that the audience will root for a character who selfishly commits fraud, risking lives and scientific progress for his own vanity.

The really scary fact is that there would be no need for a police state and segregation. The genetically enhanced would just completely dominate an open and fair competition.

[est31]

Gattaca shows a society in which eugenics, in hopes of creating a shortcut for people who have supposedly the better genes, devolved into a society where someone who did not have genetic augmentation could outcompete a whole bunch of genetically augmented people.

In the movie, either the genetic augmentation didn't work (as well) as expected, or their advantage caused the augmented people to become lazy because they got covered in undeserved status no matter how little or much they worked, as everything depended on which genes they have been "bred" for. Then someone with supposedly bad genes could run circles around them just by working hard.

Maybe some mixture, e.g. in order to protect those kids who fail at the task they have been "bred" for from considering themselves failed humans, gattaca's society adopted this model where they shower all kids in status who have the right genes. Maybe it's not the kids who are being protected but the companies selling the augmentations.

[greggman3]

100% Agree. Gattaca actually showed A SPACE PROGRAM running similar to the real space programs. I don't know about now but the first astronauts needed certain genes or they weren't allowed in. AFAIK that used to be true for fighter pilots. Want to be one you needed perfect eye sight. Bad genes = selected out.

[adabaed]

Yeah super good!!

[tekproxy]

Imagine gene therapy to fix the problems. After a few generations, many diseases will be extinct.

There's a guy on YouTube doing diy gene therapy to treat his lactose intolerance so it's not exactly science fiction.

[godelski]

I'm pretty sure the effect was temporary and he had to do it a second time. It's very important to note that this research is still very new and he was lucky that his genetic code was prime for that test. (I'm not against bio hackers btw. I think they provide a very good service though obviously more risky. No problems when that risk is on yourself but just trying to say "don't try this at home").

[jcims]

The Thought Emporium

https://youtube.com/c/thethoughtemporium

He’s got a ton of other interesting projects, the DIY gene therapy is just one that stands out because it seems so risky.

[godelski]

I'm pretty sure he had a second video on how the effects started to ware off.

[dekhn]

Note that you are literally shedding identifiable DNA from your body at all times and a truly motivated adversary would have no problem obtaining enough sample material to do high quality sequencing.

[nomercy400]

It's not the motivated adversary I am worried about, who actually has to show up where I have physically been. It is the company on the other side of the world in a country with lax legislation, profiling me based on the data I 'shed' online, like a cloud-based DNA sequencing service.

[KeepFlying]

This is my threat model for most things in life. If someone is physically targeting me, I'm fucked. I'm more worried about limiting the casual long-distance attacker since I have more ability to stop them.

If someone steals my DNA I can't stop them. But I can at least avoid being swept up in large scale DNA scanning and tracking efforts.

[ClumsyPilot]

The data monopolies and abuse originate from people giving these companies data for free. If they had to buy it, or pay goons to collect it, they wouldn't be profitable.

[russdill]

In the near future (or arguably now depending on your purpose) you don't even need that. Assuming enough of your relative's sequences are available, the probability of you having certain genes/mutations can be narrowed down so much that having your individual genome doesn't add much.

[petters]

This does not seem true? Even if the complete genome of my mother and father is known, there is still a lot of uncertainty left.

[sllabres]

Isn't that quite similar to help solving cold cases as example:

https://en.wikipedia.org/wiki/Joseph_James_DeAngelo

On April 24, 2018, authorities charged 72-year-old DeAngelo with eight counts of first-degree murder, based upon DNA evidence; investigators had identified members of DeAngelo's family through forensic genetic genealogy.

[petters]

You don't have to reconstruct the genome in order to prove relationships or find a person.

[russdill]

Lets say there's some rare genetic disorder, only a few hundreds of a percent of the population has it. If someone knows that your mother or father has it, you now don't have a few hundreds of a percent chance of having it. Depending on the disorder you having it might just be a cointoss.

[tgsovlerkhgsel]

One of the key differences is that in the case of the DNA sequencing services, you're agreeing to ToS that allow them to abuse your data (and thus indirectly the data of any of your blood-relatives), and you directly tie the data to a name and address.

[Teever]

I assume this line of reasoning is also why you don't lock your doors at night?

[shukantpal]

At scale?

[dekhn]

Sure. I've worked with and know people who could carry this out at scale, although obviously individual sample collection isn't highly scalable.

Edit: I used to help Google fund researchers like Joe Derisi and others who develop technology to do this, and some of the people I worked with in my academic career are quite good at identifying serial killers from 30 year old DNA. If you're downvoting because you think I'm making this up, you're wrong. If you're downvoting because you don't think large-scale individual detection using genetic sampling of the environment is possible, you're wrong. If you're downvoting because you think you couldn't do a whole genome sequence of an individual using a sample collected in the wild, you're wrong. If you're downvoting because you think this is a terrible idea (morally, ethically), that's fine but I didn't say anything about my own moral or ethical beliefs about this.

It's simply factually correct to say that large-scale individual sample collection (at order tens of thousands, if not hundreds of thousands of individuals in a country the size of the US) is possible. All the technology is there to do this.

[jcims]

It seems very unlikely to me that there isnt at least *some* genetic information that would be of direct value in advertising. Like if Google took five years of personalized ad performance info and filtered that through the associated individual’s known DNA to develop a predictive model.

[hourislate]

I'm curious whether a Covid PCR test could be used to sequence your DNA. Is there enough of a specimen in the process.

[josters]

I read a story[1] about a UK-based Covid testing firm who was planning on collecting and selling their customers' DNA samples.

> Its "research programme information sheet", last updated on October 21, says the company retains data including "biological samples" and "the DNA obtained from such samples", as well as "genetic information derived from processing your DNA sample ... using various technologies such as genotyping and whole or partial genome sequencing".

The policy also says Cignpost may share customers’ DNA samples and other personal information with "collaborators" working with them or independently, including universities and private companies, and that it "may receive compensation" in return.

[1] https://www.telegraph.co.uk/news/2021/11/14/covid-test-firm-...

[jcims]

This reads like the lead-in to an episode of Black Mirror.

[walterbell]

https://www.msn.com/en-us/news/us/report-la-sheriff-halting-...

> L.A. County Sheriff Alex Villanueva .. was briefed by the FBI about “the serious risks associated with allowing Fulgent to conduct COVID-19 testing,” ... the FBI advised him that information is likely to be shared with China, and that the FBI told him DNA data obtained is “not guaranteed to be safe and secure from foreign governments.”

[monopoledance]

Yes, of course there is. Just like any other medical sample you have ever given. Blood, stool, urine, swaps...

But also all the unintentional donations: Every pubic hair you lost on the toilet seat, every tampon you disposed, every bandage you ripped off and threw away, every mattress you slept on, chewing gum you've spit out, every ejaculation, every ... you get the idea.

That's why you need laws to regulate this.

[l-albertovich]

Well crap, there you go, giving the lunatics reasons not to get tested and make the whole thing worse than it already is...

I'm joking, I don't think you did anything wrong but I'd hate it if a ridiculous argument such as this example gained any traction :

Example : The government / aliens / whoever released the virus so we willingly gave them our DNA to sequence and match with our assigned ID so they can do XYZ in case the implant in the vaccine doesn't work or if we are smart enough not to get vaccinated.

Scary stuff.

[makeitdouble]

Given the past incidents, it would probably be more in line with a gov. agency getting direct access to everything that goes through a few selected labs for years/decades, so that a significant number of people geting blood/cells sampled for that period would have a high chance of passing their data though these labs at some point.

At the base of it, if the gov. of the country one lives in is the enemy, it can’t be a matter of refusing vaccines here and there, that’s not the scale they should be thinking about.

[eurasiantiger]

Absolutely.

[diplodocusaur]

I imagine one's DNA can't be too different from the cousin that agrees to share that kind of data?

[Method-X]

When I had 23andme sequence my DNA, I used a fake last name and pre-paid credit card.

[albertgoeswoof]

But if enough of your relatives get sequenced, they’ll know who you are anyway

[authed]

If your family members got it done with their real name, they will be able to create a link.

[amelius]

Uh yeah, but you logged in from an IP address which Google already tied to your real name.

[lostlogin]

It’s out the bag now - you can be identified via relatives DNA.

https://www.latimes.com/california/story/2020-12-08/man-in-t...

[dataflow]

I don't think that implies the increase in risk would be negligible, which was the parent's point.

[Gatsky]

It’s not exactly DIY but there are in theory ways to ‘encrypt’ your DNA before it gets sequenced. Something like amplifying/enzymatically modifying the DNA in a way that changes the sequence which you can undo computationally once you get the data back.

[biophysboy]

Its only valuable if somebody also interprets it for you, such as telling you whether you have a genetic predisposition for certain diseases.

[m12k]

One of the other comment threads indicates that the data, that you need to do that kind of annotation of the sequence, is to some extent available for home use as well: https://news.ycombinator.com/item?id=29695449

I'm really hoping someone will work on an open source "23andme@home" solution that ties all this together in an accessible way.

[rumblerock]

Years ago I used Ancestry, then requested the .txt file and asked them to delete it from their records. Uploaded it to run a report at https://promethease.com/ that cross-references your SNPs against the existing body of genetic research.

The results have been pretty astounding. I found markers that pointed to poor response to a specific blood thinner my grandfather was put on before he passed. Currently I'm researching the cluster of Bipolar / ADHD / SAD symptoms I experience that all seem to trace back to a certain genotype of circadian rhythm genes I have (thank you, Sci Hub). To boot, some of the studies I've come across have been done on Han Chinese populations that match my descendance.

Perhaps going too far down this rabbit hole poses a self-diagnosis risk, but the correlations to my family history and my own life experience working with doctors to diagnose and treat symptoms are pretty undeniable. And given that your run-of-the-mill psychiatrist is going to treat you off of a DSM checklist, I feel much more confident knowing there have been genomic studies to back things up, since my doctor isn't up to date on this research, and finding one that would be will be difficult and expensive. I've shared the papers with my doc and he's been supportive, sometimes I feel like I should be getting a discount on services rendered.

[ClumsyPilot]

>"poses a self-diagnosis risk"

Self-diagnosing is not the problem it is made out to be - I live with my symptoms 24/7, doctor sees me for 5 minutes. The amount of times doctors have missed fairly clear sign of trouble in my family is disturbingly high. A simple procedure, done in time, would have saved two people I know.

Unfortunately our educational system teaches you about mitochondrion, but not the practical difference between ibuprophene and paracetomol, or CRP.

[DoctorOW]

Is that not something software can theoretically provide?

[jacquesm]

Your DNA can tell you a lot about what could happen, but not about what is happening.

[refurb]

That's Prometheus, no? They got acquired however, but prior to that you could upload data anonymously and then browse the analysis. It was very rough though, just linking sequences to risk, but a lot of it was inconclusive.