The most frustrating thing about the current state of our genomic knowledge and capabilities is the gulf between them and what's actually available in the world. I feel like theorem and praxis have a vast gulf between them, and although we keep progressing on the theorem side, the praxis side has essentially zero serious investment or effort behind it.
Like we've been able to CRISPR single genes for soooo long now. But do you see any governments anywhere in the world doing a crash program to CRISPR the "you need 30% less sleep" SNP into their populations at large? Not at all. But just think of the productivity and economic growth potential! Governments should be falling over themselves to do this at scale for free to any citizen that wants it. As a parent, think of being able to give your kid literally 30% more conscious life, right off the bat, every day they're alive! It would be like saving 4k+ lifetimes every single year, just in the US.
And as far as I know, labs aren't GWAS-ing "sleeplessness" more generally, to do even better than that 30%. Certainly not in any way that's going to let it be one of the menu items from one of the gengineering / embryo selection startups like Orchid. Yet arguably, this is probably the highest possible value thing we could drive in humans with gengineering, and we could have been doing it a decade ago, at scale, and this is just *one* example.
1. Rare mutations may only have a limited role, aside from major anomalies like trisomy, mosaicism - proportional to the number of genes affected. Although the distribution of predictor weights (e.g. Figure 3 from this study https://www.nature.com/articles/s41467-019-09718-5) might suggest an extreme effect for increasingly rare variants, this is likely tempered by real-world selective pressures starting at around two standard deviations.
2. Interestingly, reaction time as shown in Figure 1 is probably the phenotype least affected by the environment and thus should demonstrate the strongest genetic influence. However, given that the observed effects were minimal, I suspect that these results, derived from the UK Biobank, might be more reflective of noise than true reality.
Hey I saw a week response by bird to your friend Uber soy could you please respond to his claims I'll quote them Uber soy I saw a response to you by Kevin Byrd can you respond to this This is some embarrassing flailing. I document several misrepresentations and inaccuracies in your video. The claims about the cause of the Flynn effect decline and the relationship of g-loaded IQ subtests and culture were just two direct refutations. You seem very confused about the fact 84% of genes are expressed somewhere in the brain at some point during development. This has no implication for racial differences unless you can specifically identify expression differences between races and their relationship to IQ. This research has not and likely cannot be done and the genetic data I presented shows there is no evidence of substantial genetic differences between races for genes associated with intelligence when you correct for biases in GWAS engage with your references the whole time, and bring up studies that address the crucial weaknesses in your cited work. It's a literature review based on some of the latest genetic studies and on economic papers that correct for the shoddy statistical analyses used in much of the IQ literature. It isn't the "sociologist's fallacy" to show that accounting for these socioeconomic differences reduces the gaps since there is strong evidence and historical documentation that these socioeconomic differences between races are not genetic themselves and again no evidence from that genetics contributes to these racial gaps. Bringing up the Coleman report is irrelevant when I present papers from this decade (not half a century ago) showing that data from 4 million students pointing toward economic inequality and segregation as driving the majority of achievement test score gap in schools. You should update your references to the proper century. Now addressing the rest of your tantrum in order: 1. Yes, correlational research is weak and needs either experimental validation or more robust methods to infer causality. 2. They are fundamentally interactions, they are not separable as genetic or environmental and they show that phenotypes can change in different environments. 3. Laughing does not refute my own published researcher showing that genes associated with intelligence do not show the patterns that would be present if natural selection were acting to make Europeans more intelligent than Africans. 4. Your evidence for dysgenics relies on faulty genetic methods prone to false-positives and from researchers with no credibility or expertise. 5. The sibling study on the Flynn is precisely the kind of well-designed study that can distinguish genetic from environmental causes and it unambiguously supports environment and precludes genetic causes. 6. Fst between dog breeds are much larger than between human populations. The paper I cited references 3-5% for human races and 27% for dogs using comparable genetic markers. 7. The distinction between within- and between group heritability is a fundamental aspect of that statistical method. Also the data I presented did show school districts where there are no racial test score gaps, a closing racial test score gap for national standardized tests, and IQ tests which show no racial gap. have to once again stress that the "g" in g-factor is not referencing genetics. genetics and the g-factor are largely unrelated thing. Also, the study about education and gender inequality is not "unknown" and uses data from three well known large studies with representative samples sizes. 9. The Ritchie and Tucker-Drob paper does not show a fade-out effect from these education gains. At least read what you try and criticize. 10. I cited papers that controlled for income and wealth and they accounted for nearly the entire gap in academic performance. 11. That paper I cited is literally the main reference in your own review paper, along with a large single-cohort study of 18,000 people showing a correlation of 0.27, which the authors settle on as the most likely value
Thanks for the interesting and informative history of polygenic scoring.
The most frustrating thing about the current state of our genomic knowledge and capabilities is the gulf between them and what's actually available in the world. I feel like theorem and praxis have a vast gulf between them, and although we keep progressing on the theorem side, the praxis side has essentially zero serious investment or effort behind it.
Like we've been able to CRISPR single genes for soooo long now. But do you see any governments anywhere in the world doing a crash program to CRISPR the "you need 30% less sleep" SNP into their populations at large? Not at all. But just think of the productivity and economic growth potential! Governments should be falling over themselves to do this at scale for free to any citizen that wants it. As a parent, think of being able to give your kid literally 30% more conscious life, right off the bat, every day they're alive! It would be like saving 4k+ lifetimes every single year, just in the US.
And as far as I know, labs aren't GWAS-ing "sleeplessness" more generally, to do even better than that 30%. Certainly not in any way that's going to let it be one of the menu items from one of the gengineering / embryo selection startups like Orchid. Yet arguably, this is probably the highest possible value thing we could drive in humans with gengineering, and we could have been doing it a decade ago, at scale, and this is just *one* example.
1. Rare mutations may only have a limited role, aside from major anomalies like trisomy, mosaicism - proportional to the number of genes affected. Although the distribution of predictor weights (e.g. Figure 3 from this study https://www.nature.com/articles/s41467-019-09718-5) might suggest an extreme effect for increasingly rare variants, this is likely tempered by real-world selective pressures starting at around two standard deviations.
2. Interestingly, reaction time as shown in Figure 1 is probably the phenotype least affected by the environment and thus should demonstrate the strongest genetic influence. However, given that the observed effects were minimal, I suspect that these results, derived from the UK Biobank, might be more reflective of noise than true reality.
Hey I saw a week response by bird to your friend Uber soy could you please respond to his claims I'll quote them Uber soy I saw a response to you by Kevin Byrd can you respond to this This is some embarrassing flailing. I document several misrepresentations and inaccuracies in your video. The claims about the cause of the Flynn effect decline and the relationship of g-loaded IQ subtests and culture were just two direct refutations. You seem very confused about the fact 84% of genes are expressed somewhere in the brain at some point during development. This has no implication for racial differences unless you can specifically identify expression differences between races and their relationship to IQ. This research has not and likely cannot be done and the genetic data I presented shows there is no evidence of substantial genetic differences between races for genes associated with intelligence when you correct for biases in GWAS engage with your references the whole time, and bring up studies that address the crucial weaknesses in your cited work. It's a literature review based on some of the latest genetic studies and on economic papers that correct for the shoddy statistical analyses used in much of the IQ literature. It isn't the "sociologist's fallacy" to show that accounting for these socioeconomic differences reduces the gaps since there is strong evidence and historical documentation that these socioeconomic differences between races are not genetic themselves and again no evidence from that genetics contributes to these racial gaps. Bringing up the Coleman report is irrelevant when I present papers from this decade (not half a century ago) showing that data from 4 million students pointing toward economic inequality and segregation as driving the majority of achievement test score gap in schools. You should update your references to the proper century. Now addressing the rest of your tantrum in order: 1. Yes, correlational research is weak and needs either experimental validation or more robust methods to infer causality. 2. They are fundamentally interactions, they are not separable as genetic or environmental and they show that phenotypes can change in different environments. 3. Laughing does not refute my own published researcher showing that genes associated with intelligence do not show the patterns that would be present if natural selection were acting to make Europeans more intelligent than Africans. 4. Your evidence for dysgenics relies on faulty genetic methods prone to false-positives and from researchers with no credibility or expertise. 5. The sibling study on the Flynn is precisely the kind of well-designed study that can distinguish genetic from environmental causes and it unambiguously supports environment and precludes genetic causes. 6. Fst between dog breeds are much larger than between human populations. The paper I cited references 3-5% for human races and 27% for dogs using comparable genetic markers. 7. The distinction between within- and between group heritability is a fundamental aspect of that statistical method. Also the data I presented did show school districts where there are no racial test score gaps, a closing racial test score gap for national standardized tests, and IQ tests which show no racial gap. have to once again stress that the "g" in g-factor is not referencing genetics. genetics and the g-factor are largely unrelated thing. Also, the study about education and gender inequality is not "unknown" and uses data from three well known large studies with representative samples sizes. 9. The Ritchie and Tucker-Drob paper does not show a fade-out effect from these education gains. At least read what you try and criticize. 10. I cited papers that controlled for income and wealth and they accounted for nearly the entire gap in academic performance. 11. That paper I cited is literally the main reference in your own review paper, along with a large single-cohort study of 18,000 people showing a correlation of 0.27, which the authors settle on as the most likely value