The Unz Review - Mobile
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 Gene Expression BlogTeasers
A Golden Age of Quantitative Genomics Here
Email This Page to Someone

 Remember My Information



=>
Search Text Case Sensitive  Exact Words  Include Comments

512px-Bellcurve.svgThis year at ASHG one of the most fascinating talks was Po-Ru Loh’s, where he reviewed the BOLT-REML method. It’s introduced in the paper, Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysis. As you likely know many diseases such as schizophrenia manifest as complex trait; that is, they’re basically quantitative in their genetic architecture. Lots of alleles in the population, at varied frequencies (e.g., it might be low frequency and large effect, or higher frequency and smaller effect). In the abstract they state that “We also observe significant enrichment of heritability in GC-rich regions and in higher-frequency SNPs for both schizophrenia and GERA diseases.” In other words, they’re getting toward the holy grail of these sorts of studies, actually fixing upon likely loci which explain the variation.

But the genesis of these methods goes back to the late 2000s, when some statistical geneticists began to synthesis the power of genomics with classical quantitative genetic frameworks and insights. Another paper which sums up this tradition is Genetic variance estimation with imputed variants finds negligible missing heritability for human height and body mass index. That is, the authors have confirmed the classical heritability estimates for height, using inferences such as twin studies, with genomic methods. Many geneticists operating just outside this field are totally unaware of the power, precision, and rapidity in advance of this set of techniques. If so, I suggest you read A Commentary on ‘Common SNPs Explain
a Large Proportion of the Heritability for Human Height’ by Yang et al. (2010)
(ungated). Here is the final paragraph:

Why have we encountered so much apparent misunderstanding of the methods and results in the human
genetics community? The core of our method is heavily steeped in the tradition of prediction of random effects and the estimation of variance due to random (latent) effects. While estimation and partitioning of variance has a long history in human genetics, in particular in twin research, the prediction of random effects is alien to many human geneticists and, surprisingly, also to statisticians (Robinson, 1991). Another reason could be the simultaneous use of population genetics and quantitative genetics concepts and theory in our paper, since these are usually applied in different applications, e.g., gene mapping or estimation of heritability. All concepts and methods that we used are extensively described in the textbooks by Falconer and Mackay (1996; chapters 1, 3, 4, 7–10) and Lynch and Walsh (1998; chapters 4, 7, 26, 27).

Please, if you read anything on this blog, read this.

 
• Category: Science • Tags: Genomics, Quantitative Genetics 
Commenters to Ignore...to FollowEndorsed Only
    []
  1. Chad says:

    On the plant side, we’ve been further blessed with the development of some amazing mapping population designs, such as the Maize Nested Association Mapping (NAM) population. http://www.sciencemag.org/content/325/5941/737.full

    They have enable a nearly all the heritability for numerous traits to be explained: http://www.nature.com/ng/journal/v43/n2/full/ng.746.html

    Genomic selection is now all the rage amongst breeders, but it still seems like there is a general sense amongst many biologists and even more so outside of biology that genetics is still struggling to explain a lot of things. Maybe I tune them out now, but it does seem like there may be a quieting down of the GWAS nay-sayers.

    Read More
    ReplyAgree/Disagree/Etc.
    AgreeDisagreeLOLTroll
    These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
    Sharing Comment via Twitter
    /gnxp/the-golden-age-of-quantitative-genomics-is-almost-here/#comment-1220320
    More... This Commenter Display All Comments
  2. Thanks for posting this, even if from 2010(!). The part about LD masking heritability was the most interesting.

    Now a bit unsure what consensous is around missing heretibility problem. Sounds like you’re saying that’s on the way out as full sequencing replaces SNP, and as newer techniques parse data better. Went to missing heritability wikipedia article, but that short article had “epigenetics” in second graf, so…., hmmm. And wikipeida refs at bottom dated to 2010, as do most web search results. Not sure if there’s a recent solid pop science article on this topic. Can’t recall seeing one. Maybe overdue.

    Read More
    • Replies: @Razib Khan
    consensus depends on the trait. that being said, for height there doesn't need to be recourse to low frequency large effect alleles. rather, it's just plain old common variants across many loci. GWAS was underpowered to pick up most of the effects, so it went missing....
  3. @Nathan Taylor
    Thanks for posting this, even if from 2010(!). The part about LD masking heritability was the most interesting.

    Now a bit unsure what consensous is around missing heretibility problem. Sounds like you're saying that's on the way out as full sequencing replaces SNP, and as newer techniques parse data better. Went to missing heritability wikipedia article, but that short article had "epigenetics" in second graf, so...., hmmm. And wikipeida refs at bottom dated to 2010, as do most web search results. Not sure if there's a recent solid pop science article on this topic. Can't recall seeing one. Maybe overdue.

    consensus depends on the trait. that being said, for height there doesn’t need to be recourse to low frequency large effect alleles. rather, it’s just plain old common variants across many loci. GWAS was underpowered to pick up most of the effects, so it went missing….

    Read More
  4. Anonymous says: • Disclaimer

    Another paper which sums up this tradition is Genetic variance estimation with imputed variants finds negligible missing heritability for human height and body mass index. That is, the authors have confirmed the classical heritability estimates for height, using inferences such as twin studies, with genomic methods.

    I don’t have access to the full paper, but in the abstract they write:

    Using the GREML-LDMS method, we estimate from 44,126 unrelated individuals that all ~17 million imputed variants explain 56% (standard error (s.e.) = 2.3%) of variance for height and 27% (s.e. = 2.5%) of variance for body mass index (BMI), and we find evidence that height- and BMI-associated variants have been under natural selection. Considering the imperfect tagging of imputation and potential overestimation of heritability from previous family-based studies, heritability is likely to be 60–70% for height and 30–40% for BMI.

    The broad sense heritability of height has been estimated at ~90% for a long time now. Are they talking about broad sense or narrow sense heritability in their paper? If the former, they’re implying that traditional methods have overestimated heritability by ~50%.

    Read More
    • Replies: @Razib Khan
    from discussion:
    Using the GREML-LDMS approach, we estimated that all the 1000 Genomes Project–imputed variants explained 56% (s.e. = 2.3%) and 27% (s.e. = 2.5%) of phenotypic variance for height and BMI, respectively. These estimates are still lower than the frequently quoted estimates of narrow-sense heritability (h2) for height (80%) and BMI (40–60%) from family and twin studies. Therefore, it seems that heritability is still 'missing'. There are two possible explanations for the remaining missing heritability. The first is that there are a large number of extremely rare causal variants, not polymorphic in the 1000 Genomes Project–imputed data or removed by imputation quality control. For example, there are >40 million and >45 million variants in the 1000 Genomes Project10 and UK10K13 databases, respectively, whereas 17 million imputed variants were used in the GREML analyses for height and BMI. Complex DNA variations such as copy number variations are also not well represented by current sequencing methods25, 26. The second explanation is that heritability is overestimated in family studies, owing to effects from factors such as common environment and assortative mating that are not properly modeled27. Results from a previous study show that the phenotypic correlation for height between distant relatives (for example, cousins) is larger than what would be expected given h2 = 0.8 under an additive model28, suggesting substantial confounding in the family-based estimate of h2 but not supporting an important role for non-additive genetic variance. A recent study29 that used extended genealogy in a large sample (n = 38,167) provided very precise estimates of the heritability for height ( = 0.69, s.e. = 0.016) and BMI ( = 0.42, s.e. = 0.018). These estimates can be regarded as the upper limits of heritability for height and BMI because shared environmental effects were not explicitly fitted in the model and these estimates could therefore still be inflated to some extent. The estimates from a within-family analysis that was free of confounding from shared environmental effects are highly consistent with heritability being 0.69 (s.e. = 0.14) for height and 0.42 (s.e. = 0.17) for BMI, but the standard errors are too large to draw strong inferences30. There has also been evidence suggesting that a population-based heritability estimate is likely to be lower than that from pedigrees31. If we extrapolate from the GREML-LDMS estimates (Supplementary Table 3) by taking into account the imperfect tagging of 1000 Genomes Project imputation (on average, across five different types of SNP arrays and four simulation scenarios, ~97% and ~68% of variation at common and rare variants is captured by 1000 Genomes Project imputation, respectively; Supplementary Fig. 4), the adjusted estimate of heritability would be 0.61 (s.e. = 0.045) for height and 0.29 (s.e. = 0.47) for BMI (see the Supplementary Note for the adjustment method). These estimates can be regarded as the lower limits of narrow-sense heritability for height and BMI. Our results suggest that heritability is likely between 0.6 and 0.7 for height and between 0.3 and 0.4 for BMI. Therefore, there is little missing heritability for these traits. These results also suggest that there is little room for the other possible sources of missing heritability (Supplementary Note).
  5. @Anonymous
    Another paper which sums up this tradition is Genetic variance estimation with imputed variants finds negligible missing heritability for human height and body mass index. That is, the authors have confirmed the classical heritability estimates for height, using inferences such as twin studies, with genomic methods.

    I don't have access to the full paper, but in the abstract they write:

    Using the GREML-LDMS method, we estimate from 44,126 unrelated individuals that all ~17 million imputed variants explain 56% (standard error (s.e.) = 2.3%) of variance for height and 27% (s.e. = 2.5%) of variance for body mass index (BMI), and we find evidence that height- and BMI-associated variants have been under natural selection. Considering the imperfect tagging of imputation and potential overestimation of heritability from previous family-based studies, heritability is likely to be 60–70% for height and 30–40% for BMI.

    The broad sense heritability of height has been estimated at ~90% for a long time now. Are they talking about broad sense or narrow sense heritability in their paper? If the former, they're implying that traditional methods have overestimated heritability by ~50%.

    from discussion:
    Using the GREML-LDMS approach, we estimated that all the 1000 Genomes Project–imputed variants explained 56% (s.e. = 2.3%) and 27% (s.e. = 2.5%) of phenotypic variance for height and BMI, respectively. These estimates are still lower than the frequently quoted estimates of narrow-sense heritability (h2) for height (80%) and BMI (40–60%) from family and twin studies. Therefore, it seems that heritability is still ‘missing’. There are two possible explanations for the remaining missing heritability. The first is that there are a large number of extremely rare causal variants, not polymorphic in the 1000 Genomes Project–imputed data or removed by imputation quality control. For example, there are >40 million and >45 million variants in the 1000 Genomes Project10 and UK10K13 databases, respectively, whereas 17 million imputed variants were used in the GREML analyses for height and BMI. Complex DNA variations such as copy number variations are also not well represented by current sequencing methods25, 26. The second explanation is that heritability is overestimated in family studies, owing to effects from factors such as common environment and assortative mating that are not properly modeled27. Results from a previous study show that the phenotypic correlation for height between distant relatives (for example, cousins) is larger than what would be expected given h2 = 0.8 under an additive model28, suggesting substantial confounding in the family-based estimate of h2 but not supporting an important role for non-additive genetic variance. A recent study29 that used extended genealogy in a large sample (n = 38,167) provided very precise estimates of the heritability for height ( = 0.69, s.e. = 0.016) and BMI ( = 0.42, s.e. = 0.018). These estimates can be regarded as the upper limits of heritability for height and BMI because shared environmental effects were not explicitly fitted in the model and these estimates could therefore still be inflated to some extent. The estimates from a within-family analysis that was free of confounding from shared environmental effects are highly consistent with heritability being 0.69 (s.e. = 0.14) for height and 0.42 (s.e. = 0.17) for BMI, but the standard errors are too large to draw strong inferences30. There has also been evidence suggesting that a population-based heritability estimate is likely to be lower than that from pedigrees31. If we extrapolate from the GREML-LDMS estimates (Supplementary Table 3) by taking into account the imperfect tagging of 1000 Genomes Project imputation (on average, across five different types of SNP arrays and four simulation scenarios, ~97% and ~68% of variation at common and rare variants is captured by 1000 Genomes Project imputation, respectively; Supplementary Fig. 4), the adjusted estimate of heritability would be 0.61 (s.e. = 0.045) for height and 0.29 (s.e. = 0.47) for BMI (see the Supplementary Note for the adjustment method). These estimates can be regarded as the lower limits of narrow-sense heritability for height and BMI. Our results suggest that heritability is likely between 0.6 and 0.7 for height and between 0.3 and 0.4 for BMI. Therefore, there is little missing heritability for these traits. These results also suggest that there is little room for the other possible sources of missing heritability (Supplementary Note).

    Read More
  6. ohwilleke says: • Website

    Naively, it seems like it should be possible to use the existing data from cases of missing heritability to determine roughly how many genes are involved in generating the missing heritability, even if we can’t find them, from the shape and parameters of a fitted bell curve of variation in the measured phenotype.

    For example, if we have a collection of studies that show that 50% of variance in extraversion is heritable, and have the raw data to back that up, and know that known genetic variants account for only 5% of the variance, we ought to be able to predict whether the number of missing variants is close to 10 to 100 to 1000 or to 10,000, because the variance in the inferred heritable portion of extraversion measurements would look different if the number of variants determining it is different.

    Do you know if any strides have been made in making estimate, not of what the puzzle pieces we are missing are, but of how many puzzle pieces we are missing, for any particular traits?

    Read More
    • Replies: @Razib Khan
    Do you know if any strides have been made in making estimate, not of what the puzzle pieces we are missing are, but of how many puzzle pieces we are missing, for any particular traits?

    yes. a friend who works in the area says height almost all captured at low thousands of SNPs. intelligence somewhat more diffuse, but probably 10 K SNPs?
  7. @ohwilleke
    Naively, it seems like it should be possible to use the existing data from cases of missing heritability to determine roughly how many genes are involved in generating the missing heritability, even if we can't find them, from the shape and parameters of a fitted bell curve of variation in the measured phenotype.

    For example, if we have a collection of studies that show that 50% of variance in extraversion is heritable, and have the raw data to back that up, and know that known genetic variants account for only 5% of the variance, we ought to be able to predict whether the number of missing variants is close to 10 to 100 to 1000 or to 10,000, because the variance in the inferred heritable portion of extraversion measurements would look different if the number of variants determining it is different.

    Do you know if any strides have been made in making estimate, not of what the puzzle pieces we are missing are, but of how many puzzle pieces we are missing, for any particular traits?

    Do you know if any strides have been made in making estimate, not of what the puzzle pieces we are missing are, but of how many puzzle pieces we are missing, for any particular traits?

    yes. a friend who works in the area says height almost all captured at low thousands of SNPs. intelligence somewhat more diffuse, but probably 10 K SNPs?

    Read More
    • Replies: @Randall Parker
    The 10k SNPs for intelligence: Do you have any idea whether most or all of them are known at this point at least as SNPs? (not necessarily yet as SNPs that influence intelligence)

    If yes, then is full genome sequencing not needed in research to track down these 10k SNPs? Do we just need lots of IQ tests combined with lots of SNP tests?

    Wikipedia says there are 85 million SNPs known so far. Is it practical to test a person for 85 mil SNPs?

    What's behind these questions: I am curious to know how far away we are from identifying most of those 10k SNPs. 5 years? 10 years?
  8. Hitler says:

    Actually there is a growing scientific argument against these heritability studies, twin studies GWAS, GCTA etc.

    These guys are basically saying what you posted in this article is nonsense.

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12060/full

    They are being taken more and more seriously now too.

    Read More
    • Replies: @Emil Kirkegaard
    For readers who just read this comment, I'd like to point out that that paper was part of a series of papers discussing the issue. One can find the other papers from this paper (one of the reply papers).

    https://www.researchgate.net/publication/272564103_Mathematical_proof_is_not_minutiae_and_irreducible_complexity_is_not_a_theory_Final_response_to_Burt_and_Simons_and_a_call_to_criminologists
  9. gwern says: • Website

    Hitler: having read that Burt & Simons link, let me just say that if that is the best that the anti-heritability people can do as of 2015, then we can rest assured indeed. The GCTA section is a joke, and they don’t even cover any of the other stuff or mention GWASes – “every GCTA study of complex social phenotypes that are relevant to antisocial behavior has identified small-to-nonexistent heritability estimates, even when finding substantial twin-study estimates within the same sample” – what planet are they on? I know, all those results, must be population stratification! No matter how homogenous the discovery sample or how many principal components used or whether the hits replicated in separate samples or separate ethnic groups…

    Read More
    • Replies: @Hitler
    Hehe. I will quote the entire paragraph you misrepresent and missunderstand(typical really).

    "Importantly, to our knowledge, every GCTA study of complex social phenotypes that are relevant to antisocial behavior has identified small-to-nonexistent heritability estimates, even when finding substantial twin-study estimates within the same sample (e.g., Trzaskowski, Dale, and Plomin, 2013; Trzaskowski et al., 2013; Viding et al., 2013). "

    You can pretend the rest of the paper and many links to their previous papers does not exist if it makes you feel better. Don't bother replying to me.

    I am interested in what Razib has to say not you.

  10. Hitler says:
    @gwern
    Hitler: having read that Burt & Simons link, let me just say that if that is the best that the anti-heritability people can do as of 2015, then we can rest assured indeed. The GCTA section is a joke, and they don't even cover any of the other stuff or mention GWASes - "every GCTA study of complex social phenotypes that are relevant to antisocial behavior has identified small-to-nonexistent heritability estimates, even when finding substantial twin-study estimates within the same sample" - what planet are they on? I know, all those results, must be population stratification! No matter how homogenous the discovery sample or how many principal components used or whether the hits replicated in separate samples or separate ethnic groups...

    Hehe. I will quote the entire paragraph you misrepresent and missunderstand(typical really).

    “Importantly, to our knowledge, every GCTA study of complex social phenotypes that are relevant to antisocial behavior has identified small-to-nonexistent heritability estimates, even when finding substantial twin-study estimates within the same sample (e.g., Trzaskowski, Dale, and Plomin, 2013; Trzaskowski et al., 2013; Viding et al., 2013). ”

    You can pretend the rest of the paper and many links to their previous papers does not exist if it makes you feel better. Don’t bother replying to me.

    I am interested in what Razib has to say not you.

    Read More
    • Replies: @Razib Khan
    I am interested in what Razib has to say not you.

    no time to comment in detail. but it's moronic to conclude that if some endophenotypes have no major heritability detected that "These guys are basically saying what you posted in this article is nonsense."

    i assume you're an idiot or just ignorant if you dismiss the work of ppl associated with visscher, wray, and price based on what some criminologists say about their phenotypes of interest. stupidity is no sin, but go away :-) [i'll ban you if you leave another idiot comment]

  11. @Hitler
    Hehe. I will quote the entire paragraph you misrepresent and missunderstand(typical really).

    "Importantly, to our knowledge, every GCTA study of complex social phenotypes that are relevant to antisocial behavior has identified small-to-nonexistent heritability estimates, even when finding substantial twin-study estimates within the same sample (e.g., Trzaskowski, Dale, and Plomin, 2013; Trzaskowski et al., 2013; Viding et al., 2013). "

    You can pretend the rest of the paper and many links to their previous papers does not exist if it makes you feel better. Don't bother replying to me.

    I am interested in what Razib has to say not you.

    I am interested in what Razib has to say not you.

    no time to comment in detail. but it’s moronic to conclude that if some endophenotypes have no major heritability detected that “These guys are basically saying what you posted in this article is nonsense.”

    i assume you’re an idiot or just ignorant if you dismiss the work of ppl associated with visscher, wray, and price based on what some criminologists say about their phenotypes of interest. stupidity is no sin, but go away :-) [i'll ban you if you leave another idiot comment]

    Read More
  12. oh, fuck, the review is so dumb, e.g.:

    Advances in molecular genomics evince that genes and environments are involved in an interpenetrating and interdependent dynamic relationship that renders the attempt to demarcate separate influences—the goal of heritability studies—illogical at both the individual and population levels.

    The burgeoning research in epigenetics, which we discussed in our original article (Burt and Simons, 2014: 248–50), provides perhaps the most visible example of these conceptual breakthroughs. Epigenetic research demonstrates that the traditional focus on the effects of inherited DNA sequences (and protein-coding genes) as causally prior and impervious to the effects of environmental influences is misguided because gene expression (whether a gene is turned on and to what degree) can be influenced by environmental factors in ways that influence behavioral development.

    don’t post here again if you think stuff like this passes muster.

    Read More
  13. Anonymous says: • Disclaimer

    From the reference links of Hitler’s article.

    http://www.ncbi.nlm.nih.gov/pubmed/24870542

    Look at the comments from the scientists below. This is no joke.

    Read More
    • Replies: @Razib Khan
    what are you referring to? don't be cryptic, it's annoying.
  14. @Anonymous
    From the reference links of Hitler's article.

    http://www.ncbi.nlm.nih.gov/pubmed/24870542

    Look at the comments from the scientists below. This is no joke.

    what are you referring to? don’t be cryptic, it’s annoying.

    Read More
  15. I saw a discussion of the paper in question by Burt and Simons elsewhere, though I don’t recollect offhand where.

    It was pointed out that the very issue of the journal in which the paper was published contained a responding article by Barnes et al., which addressed the points raised in the Burt and Simons article. There were indeed several articles intended as a back and forth on the topic of heritability.

    Here’s the Barnes paper:

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12059/abstract

    How anyone with any knowledge and a desire to get at the truth can cite one without citing the other is beyond me.

    Read More
    • Replies: @MikeS
    and @Emil

    But, Burt and Simons cite that very article in the first page of the paper Hitler posted and multiple times in other parts of it, it is a reply to that one you cite. Don't try and make it out as if Barnes and Wright had the final say.

    Barnes and Wright also fail to refute them. They literally misrepresent Burt and Simons argument and also misrepresent epigeneitcs as being about GxE and rGE. Thats total nonsense, they also skip the protien paper by the looks of it.

    Holy crap man look at the abstract of the 2014 paper by Simons and Burt. They barely mention methodology and go strait to the crux of the problem with heritability.

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12036/full

    Burt and Simons are 100% right that Barnes and Wright Misrepresent them.

  16. @Razib Khan
    Do you know if any strides have been made in making estimate, not of what the puzzle pieces we are missing are, but of how many puzzle pieces we are missing, for any particular traits?

    yes. a friend who works in the area says height almost all captured at low thousands of SNPs. intelligence somewhat more diffuse, but probably 10 K SNPs?

    The 10k SNPs for intelligence: Do you have any idea whether most or all of them are known at this point at least as SNPs? (not necessarily yet as SNPs that influence intelligence)

    If yes, then is full genome sequencing not needed in research to track down these 10k SNPs? Do we just need lots of IQ tests combined with lots of SNP tests?

    Wikipedia says there are 85 million SNPs known so far. Is it practical to test a person for 85 mil SNPs?

    What’s behind these questions: I am curious to know how far away we are from identifying most of those 10k SNPs. 5 years? 10 years?

    Read More
    • Replies: @Razib Khan
    5-10 years. i think BOLT-REML type stuff is the way to go. note that most ppl have more like 10 million SNPs. 85 mill is just total in all humans.
  17. @Hitler
    Actually there is a growing scientific argument against these heritability studies, twin studies GWAS, GCTA etc.

    These guys are basically saying what you posted in this article is nonsense.
    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12060/full

    They are being taken more and more seriously now too.

    For readers who just read this comment, I’d like to point out that that paper was part of a series of papers discussing the issue. One can find the other papers from this paper (one of the reply papers).

    https://www.researchgate.net/publication/272564103_Mathematical_proof_is_not_minutiae_and_irreducible_complexity_is_not_a_theory_Final_response_to_Burt_and_Simons_and_a_call_to_criminologists

    Read More
  18. MikeS says:
    @candid_observer
    I saw a discussion of the paper in question by Burt and Simons elsewhere, though I don't recollect offhand where.

    It was pointed out that the very issue of the journal in which the paper was published contained a responding article by Barnes et al., which addressed the points raised in the Burt and Simons article. There were indeed several articles intended as a back and forth on the topic of heritability.

    Here's the Barnes paper:

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12059/abstract

    How anyone with any knowledge and a desire to get at the truth can cite one without citing the other is beyond me.

    and @Emil

    But, Burt and Simons cite that very article in the first page of the paper Hitler posted and multiple times in other parts of it, it is a reply to that one you cite. Don’t try and make it out as if Barnes and Wright had the final say.

    Barnes and Wright also fail to refute them. They literally misrepresent Burt and Simons argument and also misrepresent epigeneitcs as being about GxE and rGE. Thats total nonsense, they also skip the protien paper by the looks of it.

    Holy crap man look at the abstract of the 2014 paper by Simons and Burt. They barely mention methodology and go strait to the crux of the problem with heritability.

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12036/full

    Burt and Simons are 100% right that Barnes and Wright Misrepresent them.

    Read More
    • Replies: @Razib Khan
    what does the protein paper have to do with this?
  19. @MikeS
    and @Emil

    But, Burt and Simons cite that very article in the first page of the paper Hitler posted and multiple times in other parts of it, it is a reply to that one you cite. Don't try and make it out as if Barnes and Wright had the final say.

    Barnes and Wright also fail to refute them. They literally misrepresent Burt and Simons argument and also misrepresent epigeneitcs as being about GxE and rGE. Thats total nonsense, they also skip the protien paper by the looks of it.

    Holy crap man look at the abstract of the 2014 paper by Simons and Burt. They barely mention methodology and go strait to the crux of the problem with heritability.

    http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12036/full

    Burt and Simons are 100% right that Barnes and Wright Misrepresent them.

    what does the protein paper have to do with this?

    Read More
    • Replies: @MikeS
    It has a lot to do with it.

    It basically shows that genes don't work in the way heritability studies assume they do. For example the same gene sequences can code multiple different proteins. So correlating things like "height increasing alleles" is retarded.

  20. @Randall Parker
    The 10k SNPs for intelligence: Do you have any idea whether most or all of them are known at this point at least as SNPs? (not necessarily yet as SNPs that influence intelligence)

    If yes, then is full genome sequencing not needed in research to track down these 10k SNPs? Do we just need lots of IQ tests combined with lots of SNP tests?

    Wikipedia says there are 85 million SNPs known so far. Is it practical to test a person for 85 mil SNPs?

    What's behind these questions: I am curious to know how far away we are from identifying most of those 10k SNPs. 5 years? 10 years?

    5-10 years. i think BOLT-REML type stuff is the way to go. note that most ppl have more like 10 million SNPs. 85 mill is just total in all humans.

    Read More
  21. MikeS says:
    @Razib Khan
    what does the protein paper have to do with this?

    It has a lot to do with it.

    It basically shows that genes don’t work in the way heritability studies assume they do. For example the same gene sequences can code multiple different proteins. So correlating things like “height increasing alleles” is retarded.

    Read More
  22. oh, you’re a total dumbass. alternative splicing has been known for a long time idiot, the molecular genetics don’t effect quantitative genetics. yeah, you’re banned for being stupid.

    this is like arguing with creationists; you don’t even know what you don’t know. i was hoping for more.

    Read More
  23. p.s. not going to give people benefit of the doubt in future comments, as they really are as stupid as they seem. i’m having flashbacks to arguing with creationists at this point.

    Read More
  24. p.p.s. having to explain why alternative splicing is not going to invalidate looking for loci associated with additive genetic variance is like having to explain why mendelian inheritance doesn’t invalidate speciation.

    Read More
    • Replies: @Anonymous
    Oh come on, epigenetics and alternative splicing are heavily involved with each other. Aka it has a massive amount to do with the argument. Even without including epigenetics, alternative splicing does not work the simple("hbd") way you think it does, thats whats in that proteome paper.

    http://www.futuremedicine.com/doi/pdf/10.2217/epi.13.32
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3038581/

    The same sequence aka allele can do different things. The same sequences can do different things even by the little different sections within the sequence.

    http://learn.genetics.utah.edu/content/epigenetics/nutrition/images/mice.jpg

    Just give up already.

  25. Anonymous says: • Disclaimer
    @Razib Khan
    p.p.s. having to explain why alternative splicing is not going to invalidate looking for loci associated with additive genetic variance is like having to explain why mendelian inheritance doesn't invalidate speciation.

    Oh come on, epigenetics and alternative splicing are heavily involved with each other. Aka it has a massive amount to do with the argument. Even without including epigenetics, alternative splicing does not work the simple(“hbd”) way you think it does, thats whats in that proteome paper.

    http://www.futuremedicine.com/doi/pdf/10.2217/epi.13.32

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3038581/

    The same sequence aka allele can do different things. The same sequences can do different things even by the little different sections within the sequence.

    Just give up already.

    Read More
    • Replies: @Razib Khan
    hey dumbshit, i'm not an HBDer. i'm a *geneticist* so stop this bullshit. heritability and quantitaitve genetics predates any understanding of DNA. just like evolution predates genetics. if you don't understand the statistics of why heritability estimates are substrate neutral, shut up. also, an allele is not a sequence. i'm being picky because the fact that you state it that way suggests you don't know what an allele is.

    (you're banned too, you're all just like creationists, talking about shit you don't know about)

  26. @Anonymous
    Oh come on, epigenetics and alternative splicing are heavily involved with each other. Aka it has a massive amount to do with the argument. Even without including epigenetics, alternative splicing does not work the simple("hbd") way you think it does, thats whats in that proteome paper.

    http://www.futuremedicine.com/doi/pdf/10.2217/epi.13.32
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3038581/

    The same sequence aka allele can do different things. The same sequences can do different things even by the little different sections within the sequence.

    http://learn.genetics.utah.edu/content/epigenetics/nutrition/images/mice.jpg

    Just give up already.

    hey dumbshit, i’m not an HBDer. i’m a *geneticist* so stop this bullshit. heritability and quantitaitve genetics predates any understanding of DNA. just like evolution predates genetics. if you don’t understand the statistics of why heritability estimates are substrate neutral, shut up. also, an allele is not a sequence. i’m being picky because the fact that you state it that way suggests you don’t know what an allele is.

    (you’re banned too, you’re all just like creationists, talking about shit you don’t know about)

    Read More
  27. i’m wondering if someone posted this thread somewhere. too many dumb uninformed comments.

    Read More
    • Replies: @Emil Kirkegaard
    Possibly. I took a look at Google, but didn't find any links from outside unz.com. Perhaps it was posted on Twitter or Facebook where Google can't see it.
  28. @Razib Khan
    i'm wondering if someone posted this thread somewhere. too many dumb uninformed comments.

    Possibly. I took a look at Google, but didn’t find any links from outside unz.com. Perhaps it was posted on Twitter or Facebook where Google can’t see it.

    Read More

Comments are closed.

Subscribe to All Razib Khan Comments via RSS
PastClassics
Are elite university admissions based on meritocracy and diversity as claimed?
A simple remedy for income stagnation
Confederate Flag Day, State Capitol, Raleigh, N.C. -- March 3, 2007
The major media overlooked Communist spies and Madoff’s fraud. What are they missing today?
The evidence is clear — but often ignored