>Such predictions, at first hit-or-miss, are becoming more accurate. One test described last year can guess a person’s height to within four centimeters, on the basis of 20,000 distinct DNA letters in a genome.
I find that comparison highly misleading, height was always a reasonably 'simple' trait with very high heritability that is (in our society) relatively uninfluenced by environment. Here's a 2009 paper showing that you can predict height reasonably by averaging the parents' height (the 'Victorian method'). https://www.nature.com/articles/ejhg20095
It's one of the few traits where you can do that well, so why would you use that as an example as to why genetics-based tests are getting better? All other phenotypes/traits are much more complex.
>When they built a predictor for coronary heart disease, for instance, Kathiresan’s team discovered that the people it predicted to have the very highest risk, the top 2.5 percent, had four times the average chance of developing clogged arteries.
So what's the base risk? If the base risk is 0.0005% then a four times higher chance is still tiny.
I think it's this paper: https://www.nature.com/articles/ng.3914
As the scientists correctly pointed out, they only looked at Europeans, so who knows how this fares with Asians or Africans (likely: not well). None of the odds ratios are above 1.1, most are around 1.05, not a very extreme change to me!
Why this matters: In countries like Australia, unlike Germany, there is no protection by the law from insurance companies who don't understand genetics. People can and have lost life insurances due to the results and questionable interpretations of genetic tests. I can only hope politicians don't read this article.
There's nothing misleading about it, nor does height have a uniquely high heritability You can also predict childrens' adult IQ and many many other traits to similar degrees based on mid-parent regression, because they are all genetically influenced.
The point of the height GWAS example is that because it is so very obviously, even to the naked eye of the layman, genetically influenced and objective (you can't argue height doesn't exist), it serves as a simple test. If GWASes can't predict height, they probably can't predict anything else. And this was exactly how height was used back in 2009 or so when the 'missing heritability' debate was still going: if all these traits are genetic and additive, why aren't the GWASes able to find anything for height (back then)? The answer turns out to simply be that there wasn't enough data and the linear model analyses were lousy. Now there is enough data and better algorithms are being applied, so as expected, height prediction works much much better.
> All other phenotypes/traits are much more complex.
No, not really. That sort of polygenicity is the norm. Look at, for example, Shi et al 2016 which attempts to model the distribution of effects & thus polygenicity of 30 human complex traits: https://www.biorxiv.org/content/early/2016/01/14/035907 . Height doesn't exactly pop out of any of the charts.
> So what's the base risk? If the base risk is 0.0005% then a four times higher chance is still tiny.
...coronary artery disease should be quite obviously a higher base rate than 0.0005%. In any case, I believe the referenced study is Inouye et al 2018, in which case the extremized prediction is ~8% vs ~30% lifetime risk: https://www.gwern.net/images/genetics/2018-inouye-cad-riskpr...
> None of the odds ratios are above 1.1, most are around 1.05, not a very extreme change to me!
C'mon people...this is not science, this is fortune telling. They are just doing a bunch of huge multiparameter curve fits, which are going to pick up signal from all sorts of things that have nothing to do with biology no matter how big their sample size is. These people will tell you with a straight face that a bunch of markers that just happen to be associated with "Northeast European ancestry" are also predictive of "Polka dancing ability," and their papers are shuffling and repackaging thousands upon thousands of nonsensical little tidbits exactly like that. Life experience doesn't just "average out." Do better.
Do you believe statistics aren't real math? I'm not sure why the complaint. Sure it may be inaccurate, but as others said, along questions like: "do you have X heritage?" are already used as predictors for certain diseases. Genetic investigations just gets a little closer to the base truth, even if no one knows exactly which genes are doing which things.
But genetics may or may not be the "base truth" in regard to disease risk. It's difficult to tease out the effects of genes as opposed to environment. When your model has thousands or millions of genetic loci this only gets harder.
Some GWAS studies thats definitely true, but ultimately if the only signal you're picking up from the genetics is someones ancestry (haplotype) and that is enough to confer accurate predictions then Im not sure what your problem is? Thats great, plenty of people in the US for example have no understanding about what their genetics might confer because they might not even know where their genetic heritage comes from.
So if I have a higher risk of heart disease than average, the advice will be to eat well and exercise. This is indistinguishable from the current guidelines without any genetic information.
No rational human being should consent to their genetic information being correlated with their IQ. This kind of data is a surefire way to convince otherwise entirely capable people that they should give up on life. It's abhorrent that anyone could conceive of trying to make money out of such a test.
Adhering to that basic advice is hard. Most people don’t do it. I’d like to see a study if adherence to diet and exercise improves with genetic predictions.
For me, it would help to have a somewhat accurate prediction tied to my Fitbit that shows me right now the benefit of my hour cardio every day. My doctor could look at my physical activity vital signs pulled from wearables during my physical and give me personalized advice “You’re going to die 6 years earlier because of your genes+your diet+your excercise with 20% margin.” Is a lot more effective than “your diet and excercise suck, it will have some unquantified negative impact. But you might be fine, who knows.”
But the idea that you have absolute risk/gains from just genetics is a little unreasonable in the world of health rather than disease. For example you might have a predisposition for being incredibly stubborn and single minded (a mental trait) and utterly overcome any slightly weak cards nature dealt you for your physique. I think almost everyone can relate to that concept. We all put effort into our weaknesses most of the time, and then become highly proficient almost in what we are worst at innately! If you were told somehow your IQ is genetically average... so? IQ is a sliding scale constantly normalised as time goes on. Average IQ now is relatively much higher than a hundred years ago. The same is true with sporting achievement we have seen linear improvement in almost every sport since records started. At some point we will reach the limitation of sports science and brush up against genetic hard realities about physical muscle or oxygen transport. But not yet, and not for most people wishing to become fit and healthy. It's a genuine mix of genetics and what you are doing with your life, not some block thats genetic and then lifestyle both influence each other. You can live your life slightly abnormally to maximise your genetics. That function is very complex and personal. To talk of average risk change with genetics is almost to completely go against the whole point of the technology!
Your reality is better than the test predicted? You beat the genetic lottery! Go show to those correlations and the whole world what you really have in you.
Your reality is worse than the test predicted? You are not making full use of your genetic potential! Work harder on yourself and you could improve tremendously. If only you had known sooner that you were born for greatness.
Sure, that's no different than astrology, but lots of people are quite happy about taking advice from astrology. Genetic testing would have the additional advantage of having at least some relation to reality.
>> Life is not binary. I will never be a pro athlete yet I can enjoy doing sports.
Sure, but what if someone did a genetic test at this point and told you that your genetics would have made you a world class something-or-other? You missed your calling, how would that make you feel? For others it might help justify a failure and cause them to give up.
For a lot of people, life is a journey of discovery and pushing limits. We've seen that telling people they're inherently good at something can make them underachieve due to reduced effort. Giving people preconceived notions of who they are can be as harmful as giving them false hope.
I love stuff like this. What we need now is a service that can estimate potential genetic outcomes when having children with a partner. This can finally open the door to making genetic information a standard in dating profiles.
Instead of choosing partners based on short term metrics and hoping for the best, people would be able to select with more confidence that their children with someone would come out the way they like.
To me this kind of tech is a great way to leave a dent in our universe. A world filled with smarter, stronger, and calmer children is a world that is eventually filled with smarter, stronger, and calmer adults, who then go on to have even better offspring and accelerate the evolution of mankind into a more civilized species. Perhaps then we can finally see the end of war and ridiculous squabbles over matter.
I personally may take a stab at creating a dating service like this if the pieces are all there someday, but I’d hope by then someone else will have beaten me to it.
people would be able to select with more confidence that their children with someone would come out the way they like.
In other words, a eugenics profile.
Why stop at dating apps? You could put the data mortgage applications, insurance profiles, and more. I can imagine some of the conversations wouldn't be too far off this:
"I love you, but with a 22% chance of our kids having brown eyes and a 56% chance of high cholesterol, I have to end the relationship."
"The bank denied the application, because you're at a higher risk for a myocardial infarction by the age of 50."
"My car insurance rates will go through the roof if I put you on my policy because you're genetically predisposed to drive more aggressively than me."
But doesn't that eugenics happen naturally already? Like people of similar race, centers of interest, character or intelligence having a higher likelihood to form a couple and if DNA has anything to do with these characters, re-enforce these traits in their children? All that will change is that people will be aware of it.
I imagine you’re trying to illustrate a problem, but the solution is very simple: You simply implement government regulations that allow for cut offs so that people born way before the eugenics era aren’t disadvantaged unfairly. You could even have the cutoffs occur in phases, such that more recently born children have more scrutiny compared to ones born in the previous generation.
It’s a sci-fi trope of a kind of human bred or engineered for ultimate survival: tall, strong, smart, brave, chiseled abs and cheekbones.
But real survival characteristics are small, hairy, lazy, put on fat easily, avoid danger wherever possible, etc. So what are you actually optimising for?
You're conflating survival with success in passing on genes.
They're very different. The most obvious example here, and one that goes all the way back to perplexing Darwin, is peacocks. Male peacocks (which is a tautology - 'female peacocks' are peahens) have absurdly enormous tails and their entire body screams 'Hey, I'm here - come eat me.' And predators do just that with much greater ease than they do for peahens. But ladies dig the peacocks with the biggest, loudest, most absurd set of feathers - and so that's what we get.
Evolution is fundamentally not so much about survival, but about passing on genes - having babies. Surviving longer generally correlates with this, but evolution would happily pick for shorter life expectancy in exchange for a greater probability of passing on those genes.
My guess is that only a few years after this tech, we'll have better tech that can give your child the desired genes no matter who your partner is. And then we'll have a new age of reproductive competition the likes of which the Earth has never seen. Not a utopia, that's for damn sure.
Actually, if you could edit genes so easily it wouldn’t matter who you mated with, so there wouldn’t be much more competition, it is merely a matter of having the money and ideas for what you want in your next child.
You're going to leave a lot of people behind with this. Which is good or bad depending on your worldview. Purging people with bad genes by profiling them could be deemed unethical though and that's exactly what will happen once you're forced to show your gene quality to everyone that's interested. It can happen indirectly through dating or directly through genocide.
There is no “purge”. People with bad genes can still have children, and will. However, over time as societal awareness grows it becomes harder. Two people with bad genes might mate less commonly, however a person with great genes may settle for someone with bad genes, producing OK children.
To me this is good, and is basically what happens now only a lot less efficiently.
Genocide is completely unnecessary, and is basically for governments with bad agendas and impatience to wait for genes to improve naturally.
I’m not going to lose sleep over debating whether I should or shouldn’t build it: I’ve made up my mind, I will build it if the necessary services all exist and no one has done it well yet.
When you’re trying to change the world there is always going to be resistance from people who want to preserve the status quo. In this case, you are that resistance.
I’m just not comfortable with this. Maybe I’m just getting old and maybe I have seen too many movies but I feel uneasy about our children and maybe us having another “grade” put on them.
I’d like to think this would be good so maybe some one will comment how this won’t eventually go too far.
If altering starts happening which I would imagine it will at one point will be no longer human (and maybe that is a good thing).
Maybe at some point like in the Altered Carbon series it won’t even matter and it will just go back to money (or maybe it will always be the case as the ultimate grade).
But these studies are merely going to give you a correlation. It will always be very hard to prove direct causality without experimentation, which on humans is not really feasible or without a full understanding of how the brain works, which we are still very far from.
The other thing is that the way I like to think of our brain is like a muscle. Our DNA drives much of the range in which we can develop our muscles, people born with a certain body type will never be an athlete, but even if you are born with good muscular capacity, a KFC-eating couch potato will never get to the olympics. What one does with this capacity matters a lot. Only science will tell but I like to think that the brains works in a similar way. Some kids will never be geniuses but there is still a wide range in which they can evolve so there is no reason to corner them in a box.
One general rule of life I've learned: it's never better to live in ignorance. If you have a chance to learn a fact and the idea of knowing it fills you with dread, you should regard the hesitation as a strong signal that you should go ahead with learning it anyway. Even if it's initially painful, the knowledge will be ultimately useful.
Scientific facts about ourselves, both general and specific, are among the most painful. We go through lives deluding ourselves. That's why modern scientific introspection is so important and why it'll ultimately lead to a better world.
If it were possible to obtain this information completely anonymously, I would jump on this chance. But I certainly don't want to give my genetic info to insurance companies, especially since I don't know what it says about me yet.
It's not my reaction I'm worried about, it's the reaction of the companies I interact with that want to make a profit off me.
Long before we'll have the technology to alter genes reliably we will be able to select embryos before implantation. This is possible even today. Better genetic testing and improvements in fertility medicine only makes it cheaper and more effective.
Here's my wild prediction for the 21st century: before 2050, China will replace its one child policy with a "you have to be at least this tall to ride" policy - only people with a genetic "score" of X or higher are allowed to reproduce. Every couple of years, X is bumped up, and the next generation is a bit smarter, prettier, and stronger than the previous one.
Not if they are also bred to be more conflict-averse. Additionally, the leadership would probably get preferential access to genetic treatments and favorable mating partners.
Great techno adventure book on this subject by http://daniel-suarez.com/ "Change Agent"
This guy has more insightful ideas in his first 20 pages than most authors have their whole lives.
Readit News
I find that comparison highly misleading, height was always a reasonably 'simple' trait with very high heritability that is (in our society) relatively uninfluenced by environment. Here's a 2009 paper showing that you can predict height reasonably by averaging the parents' height (the 'Victorian method'). https://www.nature.com/articles/ejhg20095
It's one of the few traits where you can do that well, so why would you use that as an example as to why genetics-based tests are getting better? All other phenotypes/traits are much more complex.
>When they built a predictor for coronary heart disease, for instance, Kathiresan’s team discovered that the people it predicted to have the very highest risk, the top 2.5 percent, had four times the average chance of developing clogged arteries.
So what's the base risk? If the base risk is 0.0005% then a four times higher chance is still tiny.
I think it's this paper: https://www.nature.com/articles/ng.3914 As the scientists correctly pointed out, they only looked at Europeans, so who knows how this fares with Asians or Africans (likely: not well). None of the odds ratios are above 1.1, most are around 1.05, not a very extreme change to me!
Why this matters: In countries like Australia, unlike Germany, there is no protection by the law from insurance companies who don't understand genetics. People can and have lost life insurances due to the results and questionable interpretations of genetic tests. I can only hope politicians don't read this article.
The point of the height GWAS example is that because it is so very obviously, even to the naked eye of the layman, genetically influenced and objective (you can't argue height doesn't exist), it serves as a simple test. If GWASes can't predict height, they probably can't predict anything else. And this was exactly how height was used back in 2009 or so when the 'missing heritability' debate was still going: if all these traits are genetic and additive, why aren't the GWASes able to find anything for height (back then)? The answer turns out to simply be that there wasn't enough data and the linear model analyses were lousy. Now there is enough data and better algorithms are being applied, so as expected, height prediction works much much better.
> All other phenotypes/traits are much more complex.
No, not really. That sort of polygenicity is the norm. Look at, for example, Shi et al 2016 which attempts to model the distribution of effects & thus polygenicity of 30 human complex traits: https://www.biorxiv.org/content/early/2016/01/14/035907 . Height doesn't exactly pop out of any of the charts.
> So what's the base risk? If the base risk is 0.0005% then a four times higher chance is still tiny.
...coronary artery disease should be quite obviously a higher base rate than 0.0005%. In any case, I believe the referenced study is Inouye et al 2018, in which case the extremized prediction is ~8% vs ~30% lifetime risk: https://www.gwern.net/images/genetics/2018-inouye-cad-riskpr...
> None of the odds ratios are above 1.1, most are around 1.05, not a very extreme change to me!
It's the total that matters.
https://en.wikipedia.org/wiki/Type_I_and_type_II_errors
No rational human being should consent to their genetic information being correlated with their IQ. This kind of data is a surefire way to convince otherwise entirely capable people that they should give up on life. It's abhorrent that anyone could conceive of trying to make money out of such a test.
For me, it would help to have a somewhat accurate prediction tied to my Fitbit that shows me right now the benefit of my hour cardio every day. My doctor could look at my physical activity vital signs pulled from wearables during my physical and give me personalized advice “You’re going to die 6 years earlier because of your genes+your diet+your excercise with 20% margin.” Is a lot more effective than “your diet and excercise suck, it will have some unquantified negative impact. But you might be fine, who knows.”
But the idea that you have absolute risk/gains from just genetics is a little unreasonable in the world of health rather than disease. For example you might have a predisposition for being incredibly stubborn and single minded (a mental trait) and utterly overcome any slightly weak cards nature dealt you for your physique. I think almost everyone can relate to that concept. We all put effort into our weaknesses most of the time, and then become highly proficient almost in what we are worst at innately! If you were told somehow your IQ is genetically average... so? IQ is a sliding scale constantly normalised as time goes on. Average IQ now is relatively much higher than a hundred years ago. The same is true with sporting achievement we have seen linear improvement in almost every sport since records started. At some point we will reach the limitation of sports science and brush up against genetic hard realities about physical muscle or oxygen transport. But not yet, and not for most people wishing to become fit and healthy. It's a genuine mix of genetics and what you are doing with your life, not some block thats genetic and then lifestyle both influence each other. You can live your life slightly abnormally to maximise your genetics. That function is very complex and personal. To talk of average risk change with genetics is almost to completely go against the whole point of the technology!
Your reality is better than the test predicted? You beat the genetic lottery! Go show to those correlations and the whole world what you really have in you.
Your reality is worse than the test predicted? You are not making full use of your genetic potential! Work harder on yourself and you could improve tremendously. If only you had known sooner that you were born for greatness.
Sure, that's no different than astrology, but lots of people are quite happy about taking advice from astrology. Genetic testing would have the additional advantage of having at least some relation to reality.
Life is not binary. I will never be a pro athlete yet I can enjoy doing sports.
Sure, but what if someone did a genetic test at this point and told you that your genetics would have made you a world class something-or-other? You missed your calling, how would that make you feel? For others it might help justify a failure and cause them to give up.
For a lot of people, life is a journey of discovery and pushing limits. We've seen that telling people they're inherently good at something can make them underachieve due to reduced effort. Giving people preconceived notions of who they are can be as harmful as giving them false hope.
Instead of choosing partners based on short term metrics and hoping for the best, people would be able to select with more confidence that their children with someone would come out the way they like.
To me this kind of tech is a great way to leave a dent in our universe. A world filled with smarter, stronger, and calmer children is a world that is eventually filled with smarter, stronger, and calmer adults, who then go on to have even better offspring and accelerate the evolution of mankind into a more civilized species. Perhaps then we can finally see the end of war and ridiculous squabbles over matter.
I personally may take a stab at creating a dating service like this if the pieces are all there someday, but I’d hope by then someone else will have beaten me to it.
In other words, a eugenics profile.
Why stop at dating apps? You could put the data mortgage applications, insurance profiles, and more. I can imagine some of the conversations wouldn't be too far off this:
"I love you, but with a 22% chance of our kids having brown eyes and a 56% chance of high cholesterol, I have to end the relationship."
"The bank denied the application, because you're at a higher risk for a myocardial infarction by the age of 50."
"My car insurance rates will go through the roof if I put you on my policy because you're genetically predisposed to drive more aggressively than me."
But real survival characteristics are small, hairy, lazy, put on fat easily, avoid danger wherever possible, etc. So what are you actually optimising for?
They're very different. The most obvious example here, and one that goes all the way back to perplexing Darwin, is peacocks. Male peacocks (which is a tautology - 'female peacocks' are peahens) have absurdly enormous tails and their entire body screams 'Hey, I'm here - come eat me.' And predators do just that with much greater ease than they do for peahens. But ladies dig the peacocks with the biggest, loudest, most absurd set of feathers - and so that's what we get.
Evolution is fundamentally not so much about survival, but about passing on genes - having babies. Surviving longer generally correlates with this, but evolution would happily pick for shorter life expectancy in exchange for a greater probability of passing on those genes.
Gattaca
To me this is good, and is basically what happens now only a lot less efficiently.
Genocide is completely unnecessary, and is basically for governments with bad agendas and impatience to wait for genes to improve naturally.
Maybe we can build a site like that, but you've ignored the question of whether we should build it.
I’m not going to lose sleep over debating whether I should or shouldn’t build it: I’ve made up my mind, I will build it if the necessary services all exist and no one has done it well yet.
When you’re trying to change the world there is always going to be resistance from people who want to preserve the status quo. In this case, you are that resistance.
This train of thought never terminates. People say this and then someone does the thing and then people keep saying it because it sounds good.
Deleted Comment
Dead Comment
I’d like to think this would be good so maybe some one will comment how this won’t eventually go too far.
If altering starts happening which I would imagine it will at one point will be no longer human (and maybe that is a good thing).
Maybe at some point like in the Altered Carbon series it won’t even matter and it will just go back to money (or maybe it will always be the case as the ultimate grade).
The other thing is that the way I like to think of our brain is like a muscle. Our DNA drives much of the range in which we can develop our muscles, people born with a certain body type will never be an athlete, but even if you are born with good muscular capacity, a KFC-eating couch potato will never get to the olympics. What one does with this capacity matters a lot. Only science will tell but I like to think that the brains works in a similar way. Some kids will never be geniuses but there is still a wide range in which they can evolve so there is no reason to corner them in a box.
So the idea that people in general aren't going to mistreat and over-interpret the information gathered is a little ambitious.
Scientific facts about ourselves, both general and specific, are among the most painful. We go through lives deluding ourselves. That's why modern scientific introspection is so important and why it'll ultimately lead to a better world.
It's not my reaction I'm worried about, it's the reaction of the companies I interact with that want to make a profit off me.
Despite it being fiction, it goes pretty deep into eugenics and genetic discriminationin a modern society.
https://en.m.wikipedia.org/wiki/Gattaca