Substack cometh, and lo it is good. (Pricing)

Open Thread, 6/20/2018

I try to limit the shilling, but since readers of this weblog know what this means: Helix has a sale on kits until the end of the month. The three Insitome products can be purchased for app-only cost ($29.99) as an entrance into the ecosystem. In other words, for $29.99 you can get millions of markers in your exome sequenced, as well as some positions in the rest of your genome. Helix keeps the data, but there will be no kit cost if you see a future app that you want to purchase.

* Regional Ancestry
* Metabolism
* Neanderthal

If you are not subscribed to my total feed, you might not know that my review of She Has Her Mother’s Laugh went up at National Review Online. One thing that I’ve reiterated: it’s a very big book. A single review can’t really cover many aspects of what Carl touches on, and unfortunately even during the podcast I didn’t get to chimeras.

We’re almost at 100 reviews on iTunes on 10 on Sticher. I’ll stop bugging when we reach those milestones.

It’s been obvious to me that more and more people are protecting their Twitter accounts. Instead of a conversation the platform is about snooping on blow-ups and controversies.

In rapidly adapting asexuals, the orientation of G can reflect selection rather than functional constraints. Evolutionary, population, and quantitative genetics have been long-term interests of mine, so I love stuff like this. Adding a little genomics would be even better.

The most important study of the Mediterranean diet has been retracted and The collapse of a $40 million nutrition science crusade. Basically, it seems you should just eat what you want. Probably avoid processed food and trans-fats.

Diverse haplotypes span human centromeres and include archaic lineages within and out of Africa. Centromeres are crazy.

Harvard Office of Institutional Research on Discrimination Against Asian-American Applicants. I will blog about this at some point…and I’ve been making semi-serious/jokey tweets. But people are probably not clear on my “position.” My main irritation about Harvard is the extent of the doublethink that they get away with. Drew Gilpin Faust’s pablum about inclusion and diversity being central values at Harvard is a classic “Shaggy defense”. Faust helms a finishing school for the overclass, and as such accepts the necessity of discrimination when it comes to anointing future overlords. Some hypocrisy is necessary for the ruling caste, but the juxtaposition between “neoliberal” reality of what Harvard is, and the “progressive” rhetoric of it presents to the gullible and ignorant public, is starting to become indecent.

Speaking of overclass, Tim Draper is a third generation member of the capital class, and still defends Elizabeth Holmes.

With tantalizing early results, Sarepta’s gene therapy for Duchenne raises hopes for ‘real change’.

Close inbreeding and low genetic diversity in Inner Asian human populations despite geographical exogamy.

Genetic ancestry and population differences in levels of inflammatory cytokines in women: Role for evolutionary selection and environmental factors.

Population Genomics blog. The author is Chad Rohlfsen.

Re-identification of genomic data using long range familial searches.

Inference of species histories in the presence of gene flow.

How Much Does Education Improve Intelligence? A Meta-Analysis. It’s probably time to start reviewing stuff on psychometrics. The Ian Deary book is pretty good.

16 thoughts on “Open Thread, 6/20/2018

  1. I don’t blame people for going Protected on Twitter, although it comes at a price. If you put your account on “Protected Tweets” status, you lose your blue-check verification.

    The news on diet science is rather discouraging. It doesn’t really seem to matter what diet you are doing at 14 months along, as long as you are sticking to it – and most people fail on their diets. It seems the best way to avoid being obese is simply to not get obese in the first place.

    Either that, or do bariatric surgery and hope you’re not one of the 1 in 10 or so people who still gain all the weight back (I’ve known several people who did the surgery and got it all back). Or the higher percentage of people who have complications from it.

  2. Isn’t the Inner Asia inbreeding result confounded by the authors’ choice of multiple Northern mountain isolates for their Turko-Mongolic set? These are very small populations in very low population density regions without large villages. Tajiks, used for Iranian set, inhabit more hospitable climes with far more populous villages, and some, including Yagnobi Tajiks in the study, are urban.
    The main determinants of the inbreeding patterns here may be not cultural exogamy, but population density and size of the selected ethnic groups?

  3. Too much to read! “She Has Her Mother’s Laugh” seems to be very well written (I’m about 100 pages in) but is incompatible with three-week library loan periods and long wait lists. Darnit.

    On the latest podcast, Spencer wonders about the similarity of “labneh” (Middle Eastern cheese) and Lebanon. The answer is that they come from the same Semitic root l-b-n which means “white” in Hebrew (the country is named after the white snow-capped mountain range) and has transformed to mean “milk” in Arabic (hence the connection to cheese).

  4. As a PS, I would love to get into the Insitome system, but so far as I can tell, Helix is a US-only thing.

  5. Nutrition: I have said for years that there are no bad foods, only bad people. The ongoing collapse of nutrition research has continued to prove me correct.

    Razib: is there any genomic research on obesity?

  6. The nutrition stories don’t sound that bad to me.

    The “Mediterranean diet” study wasn’t really a study of said diet. Most randomized controlled studies of diet have as intervention “tell people to do diet X” and they just don’t listen. In this case, the control group changed their diet towards the Med diet, rather than away. But there was an additional intervention, namely giving free food. So this was really a study of olive oil (and nuts). Give people olive oil and they really consume more of it, and then you can ask if they live longer. (One problem with generalizing is that this was a test of the Mediterranean diet on people who live on the Mediterranean. If you gave me a liter of olive oil a week, I wouldn’t know what to do with it. On the other hand, it’s really cheap. Why aren’t they already using all they want?)

    The lack of randomization is bad, but is it that bad? One of your links seems to claim that there is a reanalysis, but I couldn’t find it. If you do a purely observational analysis, you might worry that people who consume lots of olive oil are conscientious or obedient to authority. But people who choose to be in the free food arm of the study vs people who choose to be in the no free food arm? People who share free food with their spouse? If your study gets people to change, to actually consume more olive oil, randomization seems less important. Anyhow, they seem to say that much of the effect survives discarding the non-randomized data. (Only 14% was badly randomized? I’d think that more of the people involved would be married.)

    As for the other story, it’s a pity that they failed to do a full-fledged adversarial analysis. It looks like Taubes’s very specific theories are wrong, but that’s hardly surprising. That hardly justifies “eat anything.” Maybe if you’re institutionalized and the cafeteria determines portion sizes, then the choice of food isn’t important, but you aren’t.

  7. Dude did you know the wayback machine saved some lost haloscan comments on gnxp going like wayyyy back? I only checked one date but I found some in 2003! Why haven’t you pasted them all back into your archives???????????????????????

  8. You may have already seen this Anatoly Karlin post on strength by ethnicity. But if not, the chart on grip strength by ethnicity and sex was surprising to me. May be of interest to you.

    Look for the chart just above this quote: “This is more interesting, and really quite striking. It would appear that the weakest men (South Asians) do not have a stronger hand grip than the strongest women (Europeans).”

    Now, to be clear, the chart itself doesn’t present quite that, unless my head is tilted. But it’s surprisingly close.
    http://www.unz.com/akarlin/strength/

  9. Recently there was a lot of back and forth on the enlightenment inventing racism. Some good. Some not so good. The trigger was Pinker’s book.

    But what I didn’t see, but expected to see, was a back and forth on Karmin et al paper on Y chromosome star phylogenies. That is to say, one could write an extremely good troll arguing (anachronistically) that patriarchy was invented 4000 years ago, and we’ve been living with the consequences ever since. I could see both an alt-right (this was pure goodness) and a progressive (this was pure evil) version of that argument. So taking the science and folding it towards justifying whatever your current beliefs happen to be. Like racism and the enlightenment.

    Any thought on why this didn’t happen? Or did it happen but never caught on? Or is this something that’s lurking there ready to happen, but maybe the scientists (and better journalists) involved are being cautious enough that no ideological journalist has run with it yet?

    Obviously you may not know, but perhaps you’ve gotten plugged into that kind of discussion somewhere and have some thoughts. Might be more interesting than the matriarchal hyenas dust up. 🙂

  10. “Faust helms a finishing school for the overclass, and as such accepts the necessity of discrimination when it comes to anointing future overlords.”

    I think it is a disgrace that we allow this situation to go on. I think we should put a stop to it.

    One part of my multi-step program for cutting the academy down to size is to remove their discretion over admissions.

    I am willing to bet that most people who read this would advocate for a system of meaningful national examinations like the French Baccalauréat, the British A-levels, or the German Abitur.

    Such a system is both desirable and politically impossible.

    American parents are not prepared to find out that their precious snowflakes have skated through their inferior high schools without learning anything. And certain classes have not inculcated a love of learning among their children, nor have they called out the politicians and teachers unions who conspire against them.

    None of these people will accept the verdict of a test or system of tests. They will fall for the siren song of demagogues who claim that the system is based on sexism, intolerance, xenophobia, homophobia, islamophobia, racism, and bigotry.

    The only fair system that avoids these problems is a lottery. No one can say a lottery is unfair.

    It maybe objected that the system would destroy the exclusiveness of the selective colleges. That is true but I see it as a feature not a bug.

    The so called selective colleges have given up any pretense of educating their students and have become finishing schools where the politically correct attitudes are inculcated. They do not teach the liberal arts. They allow students to skate through with gentleman’s As and never once face an intellectual challenge.

    Once they have to face the challenge of teaching a true cross section of society, they will have find new ways of achieving distinction. Maybe they will try being good teachers.

  11. When the reformation movement came to head within the catholic church, it’s obsolescence, as far as being *the* tool of choice for moving western society forward was all but set in stone. Many lessons for humanity were propogated through their efforts, but untill this day, people are still trying to reform it, yet the preists still are still taking their rounds with the children.

    I suspect the priests of academia will still be taking their turns on their respective metaphoracal children (i.e. those easily duped by the credentialing circus) forever more into the future, yet ultimately ignsigificant in the pursit of knowledge.

    How many reaserchers in their respective fields, do you know need to have their research tools all shiny and data all easly digestable in power point slides as if it were a clensing ritual before a prayer, before they can acutally “do science”?

  12. @ Nathan, Razib,

    The male figures for India are very close to those of Poland and Sweden (Pakistan’s are outright lower, but I assume it’s some sort of weird fluke).

    Based on other data presented, Swedes are about average for North Europeans.

    Assuming that, say, Icelanders are significantly stronger than most North Europeans, as evidenced by their superlative per capita performance on all kinds of strength events, then I think it is quite feasible that the average Icelandic woman is indeed stronger than the average subcontinental man.

    Razib, I don’t know if personal impressions should count for much. I assume many of your subcontinental acquaintances are American, or at least come from the Bangladeshi upper classes. Your surname also suggests a notable and (in the past) martial family background. So they should be stronger than North European women. The idea that some ethnicities may have men who are on average no stronger than women from the strongest ethnicities was also surprising to me, but that seems to be what the data is saying…

  13. The strenght of mediterrenean diet might be that they strictly follow fixed rules. Which come from the time before chemical revolution in food, junk food industry developing a science about addiciton and so on. Skipping over a long period of very bad trends in nutrition by ignoring them.

    For example in Italy they have a custom that they they drink milk coffees only in the morning, and if you try it after 11:00 you are socially ostracized. More meaningful for health is the french cultural custom of only eating 3 meals a day. Eg. I was listening to the interview with an obesity researcher Stephan Guynet (ethnic French) and he told a story that as a child he wanted to eat an apple between a meal and his parents didn’t allow him.

    Such keeping traditional ways as a culture works now, as for some time modern food has been a decrease of quality over traditional. But with the latest trends towards healthy food, superfoods from all around the world, thousands of food supplements etc. just being traditional is suboptimal. The new trends are more science based, utilizing information how much of paricular vitamins, minerals are in what food, what are the effect of food on cancer or heart diesease probability etc. You can’t beat this something that doesn’t have use such data.

  14. @akarlin: UK Biobank’s usually good for stuff like this, huge sample sizes and comparisons of the main ethnic groups in UK, and able to use same methodology for all samples so no confound from measurement error. A quick low effort google found:

    https://www.ncbi.nlm.nih.gov/pubmed/28144980

    Estimates in Biobank participants “grip strength in South-Asian men and women was ~ 5-6 kg lower than in the other ethnic groups” (black and White European, e.g. vast majority British).

    The absolute values are behind a paywall, but are less important; if you assume something like in any of the figures you link where males+females differ by about 20 kg, its infeasible that South-Asian males could have lower grip strength than White females (because 5-6 is less than 20!).

    That said, absolute numbers (Sci-Hubbing the full paper):

    White M (n=184009, age=57): 40kg, Black M (n=3085, age=50): 40.5kg, South-Asian M(n=4457, age=52): 34kg

    White F (n=224521, age=57): 23.5kg, Black F (n=4181, age=50): 25kg, South-Asian F(n=4457, age=52): 19kg

    Given the ages, most South Asian cohort should still be first generation migrants I think.

    Strength proportional to mass more similar between than absolute levels (e.g. WM: 0.47, SAM:0.44). So seems infeasible from grip strength:body mass ratio that South Asian groups would have lower grip strength than other populations with similar body mass. Maybe some difference at tails, probably not averages.

    Incidentally, though no body height given, using the BMI equation and body mass given: https://imgur.com/a/oSoU6wL. South Asian about 2.2 inches lower stature than White majority.

Comments are closed.