5 ms·

Author here. Not sure why this turned up on HN today, but feel free to ask questions.

by dash2 1y ago

usgroup 1y ago
Does the paper claim that genetics somehow drives geographic clustering? E.g. due to emigration of those carrying certain phenotypes?
- dash2 1y ago
  I think the causality is more the other way round. Originally our title was "genetic consequences..." but we were asked to change it. If you look at the part of the paper with coalfields, UK coalfields were laid down about a million years ago, before humans ever came to the area. So that was, loosely speaking, an "instrument" for an environmental variation that might then lead to genetic variation (at area level!) But yes the key message is, there is geographic clustering at genetic level.
  - notahacker 1y ago
    I'm curious how much genetic geographic clustering could have influenced the baseline associations identified in the UK genome wide association studies you selected as a starting point (not sure they typically control for region when associating traits?), particularly with traits like conscientiousness and openness, which obviously can be influenced by polygenic factors, but are also highly influenced by regional cultural variation.
    - dash2 1y ago
      Yup, take a look at: https://www.nature.com/articles/s41588-022-01062-7.pdf https://www.nature.com/articles/s41588-022-01062-7.pdf
timmg 1y ago
I’ve been curious, for a while, about how our genes affect outcomes. There are kinda two extremes “blank-slate-ism” and “genetic-determinism”. I assume it is always some combination, with a lean in one direction or another. I know the discussions are politically fraught. But if I understand the summary, your findings lean toward the determinism side. Is that fair? How do you think of the dichotomy? Thanks!
- dash2 1y ago
  > But if I understand the summary, your findings lean toward the determinism side. Absolutely not. I don't think any serious geneticist is a genetic determinist, in fact it's hard to even know what that means... DNA without an appropriate environment is nothing but a long stringy molecule! In fact, the main impact of this paper was to help make geneticists aware that genes are confounded with geographic environments. That (plus much other research!) is one reason why researchers are now putting a lot of emphasis on family-based designs. In those, you can get truly causal estimates of the effect of a genetic variant or of a whole polygenic score, due to the "lottery of meiosis" that randomly give you genes from either your mum or dad. Now you could equally argue that the paper shows geographic environments are confounded with genes. That's true too, though sadly a lot of social science still proceeds as if it wasn't the case.
  - jjtheblunt 1y ago
    doesn't the lottery of meiosis randomly give you genes from each pair of grandparents, thus ending up with one random maternal grandparent choice and one random paternal grandparent choice at each position pair (maternal contribution and paternal contribution) in each cell (of course recursively happening during creation of haploid germ cells within each person) ?
    - dash2 1y ago
      Yes. But if you're suggesting that you could treat differences with cousins as random, the way we can treat differences with siblings, then no, because of assortative mating; e.g. if my cousin's "good genes" came from my uncle, then maybe he married my very rich aunt who left my cousin a large inheritance.
      - jjtheblunt 1y ago
        :) (wasn't suggesting that)
    - Merrill 1y ago
arethuza 1y ago
Some of the maps have places like Scotland and Wales showing up quite clearly - do you think that is real or an artefact of how the data was collected?
- dash2 1y ago
  You mean the principal components of the genetic data? That's probably real. It's well known that they cluster geographically, just because people tend to mate with other people close to them. There might also be stronger effects at borders, due to endogamy within Scots/Welsh/English in the past. There's a famous paper where they map the first two principal components of a bunch of humans and get a map of Europe out.
  - pcrh 1y ago
    >they map the first two principal components of a bunch of humans and get a map of Europe Interesting! Can you provide a link to this paper?
    - mikhailfranco 1y ago
      https://www.axiomofjoy.com/2019/02/19/genome_part_i.html https://www.axiomofjoy.com/2019/02/19/genome_part_i.html
michaelt 1y ago
> Individuals who leave coal mining areas carry more EA-increasing alleles on average than those in the rest of Great Britain. To what extent can we tell this apart from the fact almost every university student leaves their hometown, to attend university?
- dash2 1y ago
  That is an extremely good question. It's certainly part of it. I don't know if we ever divided the subjects up by university degree, but one could do that. IIRC this paper looks at Estonia and finds that even within different levels of educational attainment (e.g. university degree) people with higher EA polygenic scores are more likely to move to the capital: https://www.biorxiv.org/content/biorxiv/early/2025/05/18/2023.10.25.564036.full.pdf https://www.biorxiv.org/content/biorxiv/early/2025/05/18/202...
rawgabbit 1y ago
Please elaborate on this quote: Regional religiousness shows higher genetic correlations with personality (openness and conscientiousness) and less with SES and health traits than political preferences do, which implies additional dimensions of geographic clustering beyond high versus low SES. I interpret this is as saying while those with low social economic status vote Labour and higher SES vote Conservative, social economic status does not correlate with religiosity -- instead regliosity correlates to the BIG five personality traits. Is this correct? Can you expand on this?
lawlessone 1y ago
Would things like past class distinctions show up in data like this? Like say everyone tended to marry withing their own class at different points in history, they might tend to be more related to other people with ancestors in similar classes, than in societies that abandoned it. Not endorsing it but would be interesting if past social systems affected DNA.
- dash2 1y ago
  That's extremely likely. Indeed, for (a) ethnicity and (b) geography, it's obvious that these correlate with ancestry; people tend to marry within their neighbourhood and within their clan. That's one reason why the principal components of genetic data tend to map rather neatly to geography, and why genes can be used to make inferences about past populations.

Author here. Not sure why this turned up on HN today, but feel free to ask questions.

by dash2 1y ago

Author here. Not sure why this turned up on HN today, but feel free to ask questions.

Does the paper claim that genetics somehow drives geographic clustering? E.g. due to emigration of those carrying certain phenotypes?

dash2 1y ago
I think the causality is more the other way round. Originally our title was "genetic consequences..." but we were asked to change it. If you look at the part of the paper with coalfields, UK coalfields were laid down about a million years ago, before humans ever came to the area. So that was, loosely speaking, an "instrument" for an environmental variation that might then lead to genetic variation (at area level!) But yes the key message is, there is geographic clustering at genetic level.
- notahacker 1y ago
  I'm curious how much genetic geographic clustering could have influenced the baseline associations identified in the UK genome wide association studies you selected as a starting point (not sure they typically control for region when associating traits?), particularly with traits like conscientiousness and openness, which obviously can be influenced by polygenic factors, but are also highly influenced by regional cultural variation.
  - dash2 1y ago
    Yup, take a look at: https://www.nature.com/articles/s41588-022-01062-7.pdf https://www.nature.com/articles/s41588-022-01062-7.pdf

timmg 1y ago

I’ve been curious, for a while, about how our genes affect outcomes. There are kinda two extremes “blank-slate-ism” and “genetic-determinism”. I assume it is always some combination, with a lean in one direction or another. I know the discussions are politically fraught. But if I understand the summary, your findings lean toward the determinism side. Is that fair? How do you think of the dichotomy? Thanks!

dash2 1y ago
> But if I understand the summary, your findings lean toward the determinism side. Absolutely not. I don't think any serious geneticist is a genetic determinist, in fact it's hard to even know what that means... DNA without an appropriate environment is nothing but a long stringy molecule! In fact, the main impact of this paper was to help make geneticists aware that genes are confounded with geographic environments. That (plus much other research!) is one reason why researchers are now putting a lot of emphasis on family-based designs. In those, you can get truly causal estimates of the effect of a genetic variant or of a whole polygenic score, due to the "lottery of meiosis" that randomly give you genes from either your mum or dad. Now you could equally argue that the paper shows geographic environments are confounded with genes. That's true too, though sadly a lot of social science still proceeds as if it wasn't the case.
- jjtheblunt 1y ago
  doesn't the lottery of meiosis randomly give you genes from each pair of grandparents, thus ending up with one random maternal grandparent choice and one random paternal grandparent choice at each position pair (maternal contribution and paternal contribution) in each cell (of course recursively happening during creation of haploid germ cells within each person) ?
  - dash2 1y ago
    Yes. But if you're suggesting that you could treat differences with cousins as random, the way we can treat differences with siblings, then no, because of assortative mating; e.g. if my cousin's "good genes" came from my uncle, then maybe he married my very rich aunt who left my cousin a large inheritance.
    - jjtheblunt 1y ago
      :) (wasn't suggesting that)
  - Merrill 1y ago

arethuza 1y ago

Some of the maps have places like Scotland and Wales showing up quite clearly - do you think that is real or an artefact of how the data was collected?

dash2 1y ago
You mean the principal components of the genetic data? That's probably real. It's well known that they cluster geographically, just because people tend to mate with other people close to them. There might also be stronger effects at borders, due to endogamy within Scots/Welsh/English in the past. There's a famous paper where they map the first two principal components of a bunch of humans and get a map of Europe out.
- pcrh 1y ago
  >they map the first two principal components of a bunch of humans and get a map of Europe Interesting! Can you provide a link to this paper?
  - mikhailfranco 1y ago
    https://www.axiomofjoy.com/2019/02/19/genome_part_i.html https://www.axiomofjoy.com/2019/02/19/genome_part_i.html

michaelt 1y ago

> Individuals who leave coal mining areas carry more EA-increasing alleles on average than those in the rest of Great Britain. To what extent can we tell this apart from the fact almost every university student leaves their hometown, to attend university?

dash2 1y ago
That is an extremely good question. It's certainly part of it. I don't know if we ever divided the subjects up by university degree, but one could do that. IIRC this paper looks at Estonia and finds that even within different levels of educational attainment (e.g. university degree) people with higher EA polygenic scores are more likely to move to the capital: https://www.biorxiv.org/content/biorxiv/early/2025/05/18/2023.10.25.564036.full.pdf https://www.biorxiv.org/content/biorxiv/early/2025/05/18/202...

rawgabbit 1y ago

Please elaborate on this quote: Regional religiousness shows higher genetic correlations with personality (openness and conscientiousness) and less with SES and health traits than political preferences do, which implies additional dimensions of geographic clustering beyond high versus low SES. I interpret this is as saying while those with low social economic status vote Labour and higher SES vote Conservative, social economic status does not correlate with religiosity -- instead regliosity correlates to the BIG five personality traits. Is this correct? Can you expand on this?

lawlessone 1y ago

Would things like past class distinctions show up in data like this? Like say everyone tended to marry withing their own class at different points in history, they might tend to be more related to other people with ancestors in similar classes, than in societies that abandoned it. Not endorsing it but would be interesting if past social systems affected DNA.

dash2 1y ago
That's extremely likely. Indeed, for (a) ethnicity and (b) geography, it's obvious that these correlate with ancestry; people tend to marry within their neighbourhood and within their clan. That's one reason why the principal components of genetic data tend to map rather neatly to geography, and why genes can be used to make inferences about past populations.