Some people are rational, others less so.
There is a disturbing consequence to this truism.
The rational response to this latest development is to postpone reproduction to allow a fuller
understanding of its implications to emerge. Those of a rational persuasion will concur with this
assessment, others will agree less so. The implication of this logic should be a concern for
everyone of either mentality.
Specifically, the non-rationalists (i.e., perhaps those with an IQ less than 100) might see no particular
reason to alter their reproductive behavior as a result of this latest news. However, those with a
rationalist mindset (i.e., perhaps those with an IQ more than 100) should by logic greatly alter their
reproductive behavior. Such differential fertility would be catastrophic.
I am not aware of there ever having been an example in history in which IQ has strongly correlated with
birth year on a short time scale before. A low IQ birth cohort would be devastating for our society and the
cohort itself. Perhaps this is another rational reason to now avoid reproduction.
An author of a recent Gene Editing report from the National Academy of Medicine has essentially conceded
that an applicable CRISPR technology is now ready. All that is left is for people to talk about it.
CRISPR is no longer never, but when.
The scientific community themselves do not appear to have become aware yet of the extremely
significant implications that the polygenicity of human traits and diseases has for how CRISPR technology
as described in the urls below could be used to enable extreme modifications to human traits including intelligence.
For whatever reason these scientists believe that the CRISPR implementation described below would not enable such
a result. This is incorrect.
https://singularityhub.com/2017/08/15/g ... e-science/
http://www.nature.com/nature/journal/v5 ... 23305.html
Genesis of the 1500 IQ Humanoids
Re: Genesis of the 1500 IQ Humanoids
Stunning!!!
This is rolling out in real time!
Here is a GWAS for intelligence of 300,000 people with 200 new findings.
It is startling.
Before 2017, the research community had found only a few SNPs and almost none of them had ever replicated.
In one go, they are now reporting 200 high quality variants related to IQ.
Even with this result alone we are pushing into the territory of having the ability
to profoundly change human IQ. There are many other related GWAS such as
for Educational Attainment that have reported additional SNPs.
The below article also notes that the SNPs found have a substantial predictive ability
for Alzheimer's. It would seem that simply CRISPRing the below SNPs would
greatly reduce AD risk.
http://www.biorxiv.org/content/early/20 ... l.pdf+html
This is rolling out in real time!
Here is a GWAS for intelligence of 300,000 people with 200 new findings.
It is startling.
Before 2017, the research community had found only a few SNPs and almost none of them had ever replicated.
In one go, they are now reporting 200 high quality variants related to IQ.
Even with this result alone we are pushing into the territory of having the ability
to profoundly change human IQ. There are many other related GWAS such as
for Educational Attainment that have reported additional SNPs.
The below article also notes that the SNPs found have a substantial predictive ability
for Alzheimer's. It would seem that simply CRISPRing the below SNPs would
greatly reduce AD risk.
http://www.biorxiv.org/content/early/20 ... l.pdf+html
Re: Genesis of the 1500 IQ Humanoids
Went to dbsnp, uploaded the IQ SNP list and got a download of the batch in VCF.
Took the list and wrote a program to look for the GMAF distribution.
Looks like we are in business.
Over 200 of the IQ SNPs have minor allele frequencies over 10%.
In fact, over 100 of the SNPs have MAF above 30%.
Almost none of the SNPs are rare.
This should mean that the CRISPR technique mentioned above that is all set to launch should allow people to
homogeneize any of the good SNPs that they have amongst the over 100 with greater than 30% MAF.
What this is also confirming is that as far as this GWAS can detect, genius is not really what has been popularly portrayed at all.
There were only 2 SNPs found that had MAFs below 1% and none of these SNPs had a large effect size.
This means that high IQ is not about rare variants but instead it is about common ones.
The rational assortative mating strategy to optimize the genetic IQ of any potential offspring would involve finding a partner who was not necessarily the smartest that one could find but instead was the most orthogonal to the IQ SNPs that one had.
What this newest CRISPR technology implies is that homogeneous IQ raising alleles no longer contribute anything to the phenotype that could not be duplicated with CRISPR. One wants a partner with the most heterogeneous SNPs that the other partner genotypes with a double homogeneous IQ unfavorable combo. CRISPRing could then be used to make all the benefical variants homogeneous.
Here's the output of my program.
Each line represents some point in the distribution.
First line is the count of the SNPs starting with 0.00 and so on.
I included the 0.50 line as well, though it had to be 0 because the minor allele has to be
the smaller of the two allele frequencies.
GMAF00count 2
GMAF01count 6
GMAF02count 2
GMAF03count 6
GMAF04count 3
GMAF05count 1
GMAF06count 3
GMAF07count 3
GMAF08count 5
GMAF09count 4
GMAF10count 46
GMAF20count 50
GMAF30count 58
GMAF40count 56
GMAF50count 0
If this pattern holds for the other 10,000 IQ SNPs, then the CRISPR technology that is now read to launch would allow for an increase of human IQ to truly profound levels (possibly 25 SD). Many of these SNPs would be present with substantial MAFs in all human populations.
Took the list and wrote a program to look for the GMAF distribution.
Looks like we are in business.
Over 200 of the IQ SNPs have minor allele frequencies over 10%.
In fact, over 100 of the SNPs have MAF above 30%.
Almost none of the SNPs are rare.
This should mean that the CRISPR technique mentioned above that is all set to launch should allow people to
homogeneize any of the good SNPs that they have amongst the over 100 with greater than 30% MAF.
What this is also confirming is that as far as this GWAS can detect, genius is not really what has been popularly portrayed at all.
There were only 2 SNPs found that had MAFs below 1% and none of these SNPs had a large effect size.
This means that high IQ is not about rare variants but instead it is about common ones.
The rational assortative mating strategy to optimize the genetic IQ of any potential offspring would involve finding a partner who was not necessarily the smartest that one could find but instead was the most orthogonal to the IQ SNPs that one had.
What this newest CRISPR technology implies is that homogeneous IQ raising alleles no longer contribute anything to the phenotype that could not be duplicated with CRISPR. One wants a partner with the most heterogeneous SNPs that the other partner genotypes with a double homogeneous IQ unfavorable combo. CRISPRing could then be used to make all the benefical variants homogeneous.
Here's the output of my program.
Each line represents some point in the distribution.
First line is the count of the SNPs starting with 0.00 and so on.
I included the 0.50 line as well, though it had to be 0 because the minor allele has to be
the smaller of the two allele frequencies.
GMAF00count 2
GMAF01count 6
GMAF02count 2
GMAF03count 6
GMAF04count 3
GMAF05count 1
GMAF06count 3
GMAF07count 3
GMAF08count 5
GMAF09count 4
GMAF10count 46
GMAF20count 50
GMAF30count 58
GMAF40count 56
GMAF50count 0
If this pattern holds for the other 10,000 IQ SNPs, then the CRISPR technology that is now read to launch would allow for an increase of human IQ to truly profound levels (possibly 25 SD). Many of these SNPs would be present with substantial MAFs in all human populations.
Re: Genesis of the 1500 IQ Humanoids
Alright making some more progress.
The supplementary tables have some very interesting info.
For example of the 200+ IQ SNPs they have the Z scores and betas.
These should give us an idea of how many SNPs exert a positive or a negative influence on IQ and how much this effect is.
These are the 15 worst IQ SNPs to have:
rs2352974,rs12554512,rs2726491,rs13107325,rs9384679,rs6550835,rs12128707,rs2478286,rs12035012,rs3740422,
rs6508220,rs8054299,rs56150095, rs4852252,rs2071407
These are the 15 best IQ SNPs to have:
rs3754970,rs5750830,rs2836921,rs10028410,rs1362739,rs34701878,rs1145123,rs990747,rs2268894,rs2309812,rs34811474,
rs6019535,rs3101338,rs12928404,rs1906252
Will need to check the article to find which alleles are the ones to avoid.
Yet, none of them will have any large effect by themselves. Something of note is that these should probably be enriched with the best of all IQ SNPs. With the GWAS one should expect that
the big fish will be found first.
I uploaded the effect sizes figure. I think the reason that there is a near vertical line near 130 on the x-axis is because this is where it crosses over from positive to negative. The z scores use a more exaggerated scale than the betas. So, the z scores with an absolute value of 5 probably have a very small effect. Hmm, I guess what the boundary at -5 and +5 means is that this is the current limit of detection. It might not be possible to find any SNP with a smaller effect. Might need to wait until the sample sizes increases even more.
The supplementary tables have some very interesting info.
For example of the 200+ IQ SNPs they have the Z scores and betas.
These should give us an idea of how many SNPs exert a positive or a negative influence on IQ and how much this effect is.
These are the 15 worst IQ SNPs to have:
rs2352974,rs12554512,rs2726491,rs13107325,rs9384679,rs6550835,rs12128707,rs2478286,rs12035012,rs3740422,
rs6508220,rs8054299,rs56150095, rs4852252,rs2071407
These are the 15 best IQ SNPs to have:
rs3754970,rs5750830,rs2836921,rs10028410,rs1362739,rs34701878,rs1145123,rs990747,rs2268894,rs2309812,rs34811474,
rs6019535,rs3101338,rs12928404,rs1906252
Will need to check the article to find which alleles are the ones to avoid.
Yet, none of them will have any large effect by themselves. Something of note is that these should probably be enriched with the best of all IQ SNPs. With the GWAS one should expect that
the big fish will be found first.
I uploaded the effect sizes figure. I think the reason that there is a near vertical line near 130 on the x-axis is because this is where it crosses over from positive to negative. The z scores use a more exaggerated scale than the betas. So, the z scores with an absolute value of 5 probably have a very small effect. Hmm, I guess what the boundary at -5 and +5 means is that this is the current limit of detection. It might not be possible to find any SNP with a smaller effect. Might need to wait until the sample sizes increases even more.
You do not have the required permissions to view the files attached to this post.
Last edited by J11 on Wed Sep 06, 2017 10:13 pm, edited 2 times in total.
Re: Genesis of the 1500 IQ Humanoids
This is really great!
The Figure above was not what I was going for: I wanted a normal curve for the effect sizes.
Had to do some data massage, though looks like I got what I wanted.
(Used forward ordering for the negative effects sizes and backward ordering for the positive ones.
This allowed the Figure to turn out properly.)
Surprised that the Distribution looks so "normal".
Wasn't sure that it would out this way.
I am still not totally sure why the negatives should almost perfectly balance with the positives.
Is ying and yang really a law of nature?
The weird bridge thingy in the center results because there are few SNPs between -5 and 5 in effect size.
The big effect size variants around -10 and +10 are the ones that I included in my last post. The red dots above are the 2 rare 0.001 MAF SNPs.
The Figure above was not what I was going for: I wanted a normal curve for the effect sizes.
Had to do some data massage, though looks like I got what I wanted.
(Used forward ordering for the negative effects sizes and backward ordering for the positive ones.
This allowed the Figure to turn out properly.)
Surprised that the Distribution looks so "normal".
Wasn't sure that it would out this way.
I am still not totally sure why the negatives should almost perfectly balance with the positives.
Is ying and yang really a law of nature?
The weird bridge thingy in the center results because there are few SNPs between -5 and 5 in effect size.
The big effect size variants around -10 and +10 are the ones that I included in my last post. The red dots above are the 2 rare 0.001 MAF SNPs.
You do not have the required permissions to view the files attached to this post.
Last edited by J11 on Wed Sep 06, 2017 10:14 pm, edited 2 times in total.
Re: Genesis of the 1500 IQ Humanoids
J11 - its only going to be the wealthy and powerful that will have first (and possibly only) bite at this. Wealthy and powerful usually means these people actively strive for personal wealth and power. Think it through. The bulk of them will create a ruling elite.
Sent from my SM-G930F using Tapatalk
Sent from my SM-G930F using Tapatalk
Re: Genesis of the 1500 IQ Humanoids
This is the one that I wanted!
Might look like a bunch of wiggly smudges, though this one is speaking to me.
I hadn't put this one together without the figure.
For the SNPs with negative effect sizes all you need to do is CRISPR them out to have the full
benefit. Let's say you hit a double whammy with the worst negative effect at 0.5 at the far left of the figure
which includes 25% of the population. All you need to do is find a partner with one non-negative allele and you're all set.
Purging the negatives with the existing technology would be somewhat easy. It's not clear with the figure whether the actually allele frequencies are what is shown or (1- what is shown). Anything that was near 0 or near 1 on the y-axis would likely be heterogeneous
which would be easy to remove, the SNPs near the center would need more luck. The positive SNPs could not always be optimized with the current CRISPR implementation.
This would be very interesting. We would be able to more easily remove the negatives than add in the positives.
This would greatly skew the "normal" effect distribution.
The SNPs that would then most often distinguish people would be the positive ones.
It would be very very cool to now do a simulation with all these numbers to see what happens when you start to put all of this in a neat package.
Strategy: 1. Have a global database of 500 million potential partners.
2. Choose the most orthogonal of them.
3. Create 1000 virtual embryos.
4. Select the virtual embryo that could be CRISPRed such that the highest IQ offspring resulted by removing any
heterogeneous combos to the homogeneous genotype with two neutrals alleles and adding an increasing allele to any
of the heterogeneous genotypes.
Even with this basic technology one would see a rapid convergence to a near optimal result after only a few generations.
Would be very interesting to write such a program.
I have already run the simulation using only selecting amongst 1000 virtual embryos with two parents of average IQ.
I used very simple assumptions on how the genetics would work. My result was one would expect a 3 SD offspring under
this conditions. This is a fairly impressive outcome with almost no real technology input. Adding in the global database and
CRISPRing steps would amplify the result.
There is talk now that they are moving forward with technology that would allow for CRISPRing at the gamete stage. This might further enhance the potential for an IQ upgrade.
Might look like a bunch of wiggly smudges, though this one is speaking to me.
I hadn't put this one together without the figure.
For the SNPs with negative effect sizes all you need to do is CRISPR them out to have the full
benefit. Let's say you hit a double whammy with the worst negative effect at 0.5 at the far left of the figure
which includes 25% of the population. All you need to do is find a partner with one non-negative allele and you're all set.
Purging the negatives with the existing technology would be somewhat easy. It's not clear with the figure whether the actually allele frequencies are what is shown or (1- what is shown). Anything that was near 0 or near 1 on the y-axis would likely be heterogeneous
which would be easy to remove, the SNPs near the center would need more luck. The positive SNPs could not always be optimized with the current CRISPR implementation.
This would be very interesting. We would be able to more easily remove the negatives than add in the positives.
This would greatly skew the "normal" effect distribution.
The SNPs that would then most often distinguish people would be the positive ones.
It would be very very cool to now do a simulation with all these numbers to see what happens when you start to put all of this in a neat package.
Strategy: 1. Have a global database of 500 million potential partners.
2. Choose the most orthogonal of them.
3. Create 1000 virtual embryos.
4. Select the virtual embryo that could be CRISPRed such that the highest IQ offspring resulted by removing any
heterogeneous combos to the homogeneous genotype with two neutrals alleles and adding an increasing allele to any
of the heterogeneous genotypes.
Even with this basic technology one would see a rapid convergence to a near optimal result after only a few generations.
Would be very interesting to write such a program.
I have already run the simulation using only selecting amongst 1000 virtual embryos with two parents of average IQ.
I used very simple assumptions on how the genetics would work. My result was one would expect a 3 SD offspring under
this conditions. This is a fairly impressive outcome with almost no real technology input. Adding in the global database and
CRISPRing steps would amplify the result.
There is talk now that they are moving forward with technology that would allow for CRISPRing at the gamete stage. This might further enhance the potential for an IQ upgrade.
You do not have the required permissions to view the files attached to this post.
Last edited by J11 on Wed Sep 06, 2017 10:44 pm, edited 2 times in total.
Re: Genesis of the 1500 IQ Humanoids
What is all so interesting about this is that this line of thought has never been possible before.
When two very intelligent parents have children, it is expected that their children would regress to the mean.
There has never been any way to actually prevent this from happening.
Now that the SNPs are being discovered one could be quite rational about partner selection.
With a database of hundreds of millions of potential partners a supercomputer could find the perfect
genetic match to enhance intelligence in offspring. Would be interesting to know how large an effect this along
with CRISPR could have on increasing IQ.
This new research will present various methods of varying technological sophistication to increase the IQ of the next generation.
When two very intelligent parents have children, it is expected that their children would regress to the mean.
There has never been any way to actually prevent this from happening.
Now that the SNPs are being discovered one could be quite rational about partner selection.
With a database of hundreds of millions of potential partners a supercomputer could find the perfect
genetic match to enhance intelligence in offspring. Would be interesting to know how large an effect this along
with CRISPR could have on increasing IQ.
This new research will present various methods of varying technological sophistication to increase the IQ of the next generation.
Re: Genesis of the 1500 IQ Humanoids
Sure the real politik might be somewhat interesting, yet I think here again this is trying to probe the collective consciousness
of an on average abysmally unintelligent primate species and intuit how it will respond to this development. Once the 100 SD
arrive, they would quickly realize that humans are extremely limited in comprehension. This would give them overwhelming
legitimacy to rule and then they would likely do the wisest thing and create a mass roll out of a high IQ species.
A ruling elite that tried to monopolize this development by excluding others from taking advantage of it would be called out as behaving in an grossly unintelligent manner. The ruling elite would lose all claims of intellectual legitimacy. Elite power is based on
exhibiting high intelligence not very low intelligence.
The new technology will allow for a very accurate measurement of intelligence based on SNP data alone. Elite members who wanted to pursue an unintelligent strategy to maintain a power monopoly might at some point be required to prove their authority based on a genetic IQ test. Society might now be heading towards a new form of democracy based upon a requirement of leaders to demonstrate extreme polygenic IQ scores.
The discussion could then shift to what some of the most extremely intelligent members of the community thought
was the best policy choice. The people at some level would have a veto even to this arrangement, as even at the extremes
human IQ probably maxes out at only about 10 SD.
of an on average abysmally unintelligent primate species and intuit how it will respond to this development. Once the 100 SD
arrive, they would quickly realize that humans are extremely limited in comprehension. This would give them overwhelming
legitimacy to rule and then they would likely do the wisest thing and create a mass roll out of a high IQ species.
A ruling elite that tried to monopolize this development by excluding others from taking advantage of it would be called out as behaving in an grossly unintelligent manner. The ruling elite would lose all claims of intellectual legitimacy. Elite power is based on
exhibiting high intelligence not very low intelligence.
The new technology will allow for a very accurate measurement of intelligence based on SNP data alone. Elite members who wanted to pursue an unintelligent strategy to maintain a power monopoly might at some point be required to prove their authority based on a genetic IQ test. Society might now be heading towards a new form of democracy based upon a requirement of leaders to demonstrate extreme polygenic IQ scores.
The discussion could then shift to what some of the most extremely intelligent members of the community thought
was the best policy choice. The people at some level would have a veto even to this arrangement, as even at the extremes
human IQ probably maxes out at only about 10 SD.
Last edited by J11 on Wed Sep 06, 2017 10:32 pm, edited 1 time in total.
Re: Genesis of the 1500 IQ Humanoids
Strategically I think it would be such a strong tactic for there to be a global fertility boycott.
Human fertility is already desperately low.
In some nations severe demographic collapse is already underway.
Further it is recognized that those of higher intelligence have reduced their levels to that
below of other intelligence groups.
Under these circumstances, a rapid capitulation by those who might want to monopolize this new
technology would then not be unexpected. It is all too obvious how devastating a deliberate removal
of g would have on the world. In such a circumstance a ruling class would be on a rapidly sinking
ship. It would soon be apparent even to them that controlling a world of poverty, misery and ignorance
would be no great reward. Truly intelligent people would want to rule a world of wealth, happiness and
extreme intelligence.
Human fertility is already desperately low.
In some nations severe demographic collapse is already underway.
Further it is recognized that those of higher intelligence have reduced their levels to that
below of other intelligence groups.
Under these circumstances, a rapid capitulation by those who might want to monopolize this new
technology would then not be unexpected. It is all too obvious how devastating a deliberate removal
of g would have on the world. In such a circumstance a ruling class would be on a rapidly sinking
ship. It would soon be apparent even to them that controlling a world of poverty, misery and ignorance
would be no great reward. Truly intelligent people would want to rule a world of wealth, happiness and
extreme intelligence.
