A new type of genetic profiling
economist.com
economist.com
- Hiring childhood development specialists
- Pre-K programs
- Kindergarten enrollment
- Providing kids with excellent nutrition
- Providing kids with a home that is safe and free from abuse
Every single factor listed above is demonstrably correlated with lifetime intelligence, body development, mental/physical health and athleticism. Is it fair that some kids get to grow up in a safe and loving environment, while other kids worldwide grow up in malnourished or abusive households. Is it fair that some kids get best-in-class schooling and attention thanks to rich parents, while others are left watching TV all day. If we're going to ban genetic profiling because it is unfair, should we be banning all of the above as well.
Instead of trying to pull everyone down to the lowest common denominator, perhaps we should instead focus on pulling everyone up to their highest potential. I for one welcome a world of universal access to genetic improvements, and generations of kids who are healthier and smarter than we can ever hope to be.
There is a potential for human mono-culture just like in agriculture.
The stuff that's scary is not that there is universally accessible genetic improvements.
The fear is that these improves are not universally accessible.
Demanding perfection or nothing is perhaps only an ideal approach in scenarios where nothing is, in fact, an acceptable answer. In a world where Tay-Sachs exists, I think that would be an ethically abominable outcome.
In medicine we already see this where companies continually make new drugs that aren't always much better than the old one, but is 'better enough' that their salespeople convince doctors to use that one instead of the old one that has a generic.
What could we do to ensure that from day 1 this is universally accessible? This is VERY dangerous territory, we are talking about potentially bifurcating the human race here.
I understand if some take a different perspective on the matter.
But the whole debate around genetic engineering is that it isn't marginally better. The fear is that it is so supremely powerful it will dwarf all other interventions.
And I don't think ladder kicking is particularly true. Once education was only for the elite. But we have seen it get more and more popular, to the point it's about as universal as possible now.
Today's genetically improved babies could be sub-optimal by the time they hit kindergarten and just plain old obsolete before they go into high school. And the bad news is, I could see the rate of improvement increasing over time. So that material improvements are being made year to year.
Doesn't bode well for when you're going to look for a job at 29 and you have to compete with 28 year olds. Even worse if you're 29 and you have to compete with 25 year olds. And heaven help you if you're 29 and you have to compete with a 20 year old.
Someone of mediocre intelligence is not harmed by being in a society of people much smarter than him. On the contrary - he benefits greatly because smarter societies are safer, fairer, and take care of citizens (including the dumb ones).
This is revealed by real-life preferences: People from low IQ countries consistently try very hard to migrate to societies of smarter people. People from low IQ regions try to get into neighborhoods and schools of higher IQ people.
It's like complaining that Einstein existed because my grandpa didn't get to invent relativity. Well, my grandpa was never going to invent relativity. The fact that Einstein existed just means that the rest of us now get to enjoy the positive externalities created by his ability.
It's true that if everyone else is smarter, your relative social status would be lower. But this isn't an argument that scales up, since relative social status is in fixed total supply anyway.
Or am I making a mistake by trying to apply logic here?
Mother Nature has no logic. Or rather, if she has logic, it's the logic of simple chromosomal propagation on galactic timescales. She just keeps propagating different combinations and permutations of genes blindly. Then, for instance, if some cataclysmic environmental change happens she ruthlessly culls all of the gene combos that are not able to survive the new environmental reality. Or it could be a new predator. Or a new asteroid strike. Or a new bacteria. Or what have you.
Point is, in the case of our species, she doesn't make people "dumber", or "smarter", or "faster", or "stronger", or "slower", or "darker". She just blindly makes people. Of as many kinds as she possibly can. So that when the change roulette wheel spins, she's already laid bets on every number on the table, black and red alike. Thus giving her the best chance at a win. If I believed in this sort of thing, I might say that Father Time has taught Mother Nature that the shotgun approach is the best survival bet she can make. (And he probably taught that lesson the hard way.)
Contrast that blind propagation strategy against what we would be doing, which is laying odds on certain, but not all, numbers on the hypothetical roulette table. You can kind of see why people might question the wisdom of such a targeted strategy.
> - Kindergarten enrollment
> Every single factor listed above is demonstrably correlated with lifetime intelligence
I thought that pre-k and kindergarten programs (along with essentially all education) were not correlated with IQ when measured later in life? Part of how most "shared environment" influences decrease as people age. (The two remaining factors would be "nonshared environment" and "genetics", each at ~50%)
There is a correlation between "years spent in school" and IQ, but that's typically thought to go the other way. (Higher IQ causing more years in college - grad school, PHD programs, etc)
EDIT: I should add that there are still correlations that go the other way - more intelligent, wealthier parents are more likely to enroll their kids in pre-k programs and/or kindergarten, and are also more likely to have more intelligent, wealthier children. But intervention tests - where entire regions are enrolled in preschool, for instance - don't show the improvement you'd expect.
HAH. Not even close, and not never, probably. That's what people are concerned about.
The government should just step in and make this available to everyone that wants it. It would lead to a rapid improvement in the quality of its populace.
For once, these items are not eugenics, and people try to provide their kids with what's best for them.
You can deplore that it's easier for rich parents than for poorer ones, and we should try to improve it; but it's really different from selecting genetics traits before implanting an embryo.
"The twist would be to decide, on the basis of their dna, which of a group of available embryos should be implanted and brought to term."
In this case you are deciding which group of cells to implant. In your examples you are effecting a specific person. Not choosing them out of a group and discarding the others.
Personally I don't know. But your comparison isn't fair.
https://slatestarcodex.com/2018/11/13/preschool-much-more-th...
... which hardly seems cost-effective.
https://www.gwern.net/Embryo-selection
It's long, but there's a concise summary at the top that's definitely worth a read.
Embryo selection works best when there are a very small number of genetic variants that you want to select over. Want a child to not be born with Tay-Sachs disease? Can do! Want a child to be smarter or better-looking? That's going to be a lot harder, since those traits are influenced by a huge number of genetic variants, each with small effect, and we've only got a handful of embryos to choose between, which limits the amount of influence we can have here.
So, doing embryo culling works for specific, simple, detectable conditions, but the idea that it could be anything else without a lot of knowledge we as a species don't have is delusional at this point in time.
Individual people can always disagree, but average opinion can be quantified. Take a bunch of pictures, throw them on Tinder and see how many matches they get.
Although for men height is often a bigger factor than small variations in looks. Allow embryo selection for 6'+ sons, that's where the real money will be!
Yes, but height is influenced by hundreds of genes, each with a minuscule influence, which will make it all but impossible to create a model that predicts height accurately.
Additionally, since all the embryos you can select from are from the same two parents, models created from population samples are going to be a lot less useful.
So I guess you won't be able to select for tall sons, at best you might be able to pick the larger of two embryos with slightly higher than random chance.
Not at all. The current PGSes for height already predict ~40% variance (or r=0.63): see https://www.biorxiv.org/content/early/2017/09/19/190124 or https://www.biorxiv.org/content/early/2018/07/24/375337
> Additionally, since all the embryos you can select from are from the same two parents, models created from population samples are going to be a lot less useful.
No? It's all genes. The genes within a family are the same as across the population. The PGS is reduced by half* simply because you are comparing half-related embryos (siblings), so they will share many of the same genes and there are no differences to select on, but that's the nature of it. You could do selection on unrelated embryos if you really wanted to, but few people seem to want to...
* between-sibling PGS reduction is sometimes more depending on other factors like 'nature of nurture'; this is something that affects EDU more than IQ, but other traits like BMI or height appear to not be affected
> So I guess you won't be able to select for tall sons, at best you might be able to pick the larger of two embryos with slightly higher than random chance.
It'd actually be 65% with the current height PGSes (upperbounded at perhaps 71% as WGS can recover the full heritability: https://www.biorxiv.org/content/10.1101/588020v1 ).
> It'd actually be 65% with the current height PGSes (upperbounded at perhaps 71% as WGS can recover the full heritability: https://www.biorxiv.org/content/10.1101/588020v1 ).
Wouldn't knowing 65% of the variance let you pick the best of two more than 75% of the time? (my intuition would be 50 + 65/2 %, but I'm probably wrong there)
I do have an open StatsExchange question on this: https://stats.stackexchange.com/questions/427292/probability... But to calculate the probability, I just Monte Carlo it.
https://infoproc.blogspot.com/2017/09/accurate-genomic-predi...
> One of the key problems is that polygenic scores are the sum of many individual small genes’ effects and form a normal distribution, which is tightly distributed around a mean; the further into the tail one goes, the larger a sample it requires to realize a gain—to put it another way, if you have 10 samples, it’s easy (a 1 in 10 probability) that your next random sample will be the largest sample yet, but if you have 100 samples, now the probability of an improvement is the much harder 1 in 100, and if you have 1000, it’s only 1 in 1000, and worse, if you luck out and there’s an improvement, the improvement is ever tinier.
No, it's not, because you are selecting on genotype rather than phenotype.
Embryo selection only works when there is a simple connection between genotype and phenotype (eg. what type of earlobes do you prefer). For complex traits like "intelligence", selecting based on genotype is not going to work. To much noise, too little signal.
Fisher showed in 1918 you can perform selection even if there were an infinite number of loci involved, see https://en.m.wikipedia.org/wiki/Infinitesimal_model
When we compare to Pre-K programs and other current enhancement that can be bought with money, they are no guarantee that the enhancement will go to the next generations, unless the family can keep paying for those things in the future. So the issue is the starting point of the human life.
Today, we still can believe that all human are born "equal"; but the moment a whole set of human beings are enhanced on a genetic level, a lot of our moral values will be changed forever.
No longer we can say that "All human beings are born free and equal in dignity and rights." if that happens, it will upend things like the "Universal Declaration of Human Rights" and many other concepts that today's society relies on to organize itself. Imagine to live in a society where you are genetically "inferior" and there is no way you can "change" or upgrade who you are.
If someone discovered a reliable way to detect the future sexuality of a child then those efforts would suddenly be contradicted, because 1) it would be harder to argue that it's not just a condition, and 2) no matter how controversial, there would be a demand for such screenings, and the LGBTQ-community would most likely interpret it as an existential threat.
Obviously, I chose sexuality as an example as it's an easy metaphor that most people get, but this reasoning can be applied to quite a few other things that are (possibly) caused by genetical factors and where there's a community involved. As an example, I've heard people mention an on-going "genocide" of people with Downs syndrome, simply because pre-natal screenings are quite effective at detecting it and that parents therefore terminate the pregnancy instead of having a child with Downs.
EDIT: Simply put: If you start screening for X, thereby allowing it to be prevented or treated in any way, then you imply that the X is something negative. People with X might take offence to that.
We can. Humans are equal in dignity and rights. That doesn't say anything about genetics.
The Egalitarian Fallacy
https://link.springer.com/article/10.1007/s11406-019-00129-w
Eventually, of course, the embryos could also be modified by CRISPR or some similar technique, and those for which the genetic modification was successful would then be chosen for implantation.
Outlawing this sort of thing is likely to have two effects:
- ordinary folks will do it illegally in back street genetic engineering operations, and
- rich folks will engage in medical tourism to some place set up to cater to their needs - a Cayman Islands of genetic engineering.
PS - I'm not sure what the problem with cloning is, since identical twins, triplets, ... are naturally occurring clones.
The minimum possible effect time, if they have no descendants.
This article chose to use the word "profiling"
People choose to refrain from other loaded words like "eugenics"
but this is exactly what people do pursuing others that they consider attractive. nobody would say they are, with the idea of eugenics reserved for 1930s state sanctioned medical pogroms, but their actions of "choosing inheritable traits" undermine that.
takes tin foil hat off
Perhaps it could be thought of as positive & negative eugenics (like positive & negative rights)? E.g. killing people and telling them who they are allowed to marry and whether they can have kids is negative (taking away), while growing embryos free of genetic diseases is positive (giving)?
Just rolling the idea around in my head. Of significance to me, if we had the capability to completely eliminate diseases like tay-sachs in a positive way (e.g. embryo selection), it might be inhumane not to.
does that mean you think that we "are born into this world" (nevermind if we are born with good or bad fortune)?
I'm just nitpicky about the tacit implications of that common phrase "born into the world"
Also, the understanding isn't there yet to detect casual SNPs. We know roughly where these SNPs are, but not which ones they are. We can tell that if you have this region of the genome you may be taller, but along with it might come linked diseases.
So no, no designer babies yet, but some kinds of selection are possible in theory.
In real life yes, because people see your height relative to others.
In online dating it's all in comparison to nice round numbers like 6'. 80% of men are under 6' in real life, because the literal average is below that. But online 6' becomes the magic marker and anyone shorter is judged "below average".
There are already successful GWASes for attractiveness: https://journals.plos.org/plosgenetics/article?id=10.1371/jo... This should not be shocking, since a lot of it is just tapping into weight/obesity phenotypes, which are very obviously heritable.
Symmetry would be an easy "attractive" metric, but these are all complex traits that we aren't really close to narrowing down the mechanism for.
As one would be expected from GWASes with current sizes and noisy measurements. Yet, that's very different from 'never'.
> Yes it passes GWAS threshold but a lot of associations do that are later found to be false positives
That's not true. Genome-wide statistically-significant hits have excellent replication rates, when calculated properly.
There's snps for unattractive, eg rs1934328, involved in ~40% of all cases of nonsyndromic cleft lip with or without cleft palate.
"We were just wondering if it's good to leave a few things to chance?"
"We want to give your child the best possible start. Believe me, we have enough imperfection built in already. Your child doesn't need any more additional burdens. Keep in mind, this child is still you. Simply the best of you. You could conceive naturally a thousand times and never get such a result."
-- Gattaca
However -
The story of the film tells us time and time again that Vincent -isn't- able to compete and measure up to the standards needed for space travel. He fails his cardio tests, he can't see well enough to cross the street. But his vanity and pride causes him to cheat the system to make it seem like he passes all the tests. He has the heart of a fighter figuratively but not literally and I always imagined after the end of the movie that his actual faults would come to light and it would cost someone their lives in space.
What if he lost a contact lens in space? They don't have more. What if there was an emergency and they needed someone with an expected amount of endurance to help save everyone? Gattaca for me always told a different, darker story. I think the story they wanted was a caste story - if he was just as able and strong but beat out the genetically superior people on their terms then that would have been the story they were trying to tell. In the end, they told the story of a clever cheater who put his own wants above those in a delicate mission and it's left to a guess if that cost lives or no.
During a talk with another person on this, he supposed that this is why Jerome incinerates himself at the end. He needed a lowly sort as such as Vincent to fulfill his own dream, and in the end just could not reconcile how he may have just destroyed and polluted the whole of his pure society for his own selfish reasons.
To have Vincent be just as genetically able and strong would defeat the whole point. Not all men are born equal, life is unfair but we should all aspire to reach for our dreams despite our limitations. If we take things to their materialistic dead-end we get a single man optimally selected for a single task from his birth unto his death like a machine with no room for error, or chance, or humanity.
Enough of those people typically read and comment on the article, often quoting the parts that they reply to, to make the comments interesting and informative to those of us who have not read the article.
Also, HN comments often go off on a tangent, and those tangents are often the best part of the discussion.
If a person can't read the article and the comments aren't useful without having read the article--my recommendation is to move on to another story. HN has sufficient volume that there is no way most people have time to read more than a fraction of it anyway.
Finally, if you notice that there are a lot of submissions that you really do want to read coming from the same paywalled source, that's a good indication that maybe that is a site you should subscribe to, either directly or as part of some aggregator like Apple's News+ service.