3 ms·

This makes little sense. Sure, maybe alcohol was invented first in China (that's the earliest we have found anyway). But we have found usage in ancient Persia

by nl 5y ago

This makes little sense. Sure, maybe alcohol was invented first in China (that's the earliest we have found anyway). But we have found usage in ancient Persia from less than 2000 years after China. If this gene is highly selective, why wasn't it also equally selective outside East Asia? A more likely explanation is that it is a random variation and provides little to no evolutionary benefit.

nsomaru 5y ago
If it was random, why did it become dominant if it provided no evolutionary advantage? There’s variations going on all the time, maybe they’re random, maybe they’re not. But the ones that stick in a population are the ones that get selected by outcompeting hosts without the variation.
- AstralStorm 5y ago
  You could probably also ask why Temujin had many children. Founder effects are a thing and a sufficient explanation. They hold even for relatively detrimental defects, and likely hold even better for something as silent as acetaldehyde metabolism.
- nl 5y ago
  > There’s variations going on all the time, maybe they’re random, maybe they’re not. But the ones that stick in a population are the ones that get selected by outcompeting hosts without the variation. This is only true if there is correlation between the variation and survival in terms of breeding. A good example is the high incidence of Haemophilia B in European royalty[1]. It certainly doesn't give any selective advantage, but the fact that royalty are more likely to get medical care means it doesn't have any selective disadvantage either. There are plenty of other random variations that aren't selective. Eye colour, male pattern baldness, the range of earlobe attachedness[2] are all variations that don't have any selective pressure and so their appearance in various geographic areas is down to history rather genetics. Edit: Another comment led me to a paper arguing compellingly that this gene provided some protection against diseases carried in mouldy rice, and that was selective in China. See further discussion here: https://news.ycombinator.com/item?id=27705633 https://news.ycombinator.com/item?id=27705633 [1] https://en.wikipedia.org/wiki/Haemophilia_in_European_royalty https://en.wikipedia.org/wiki/Haemophilia_in_European_royalt... [2] https://en.wikipedia.org/wiki/Earlobe#Genetics https://en.wikipedia.org/wiki/Earlobe#Genetics
dynm 5y ago
These mutations do have a lot of spread outside of China. For example Fig. 1 of this paper (http://www.medgenetics.ru/UserFile/File/Borinskaya%202009%20AJHG%20Distr%20ADH1B%2047His.pdf http://www.medgenetics.ru/UserFile/File/Borinskaya%202009%20...) shows a high incidence of ADH1B in the Middle East. Is it really required that these genes are equally selective everywhere? If the adaptation theory were true, that would depend on how widely available alcohol was, as well as how harmful it is to lose the extra calories from rotting fruit, etc. No particular reason to think it would be equal everywhere.
- nl 5y ago
  Thanks for this. That paper led me down a rabbit hole and I ended up with this: Evidence of positive selection on a class I ADH locus It seems unlikely that the selection was recent and associated with alcoholism, the modern phenotypic manifestation of the polymorphism.... Mycotoxins, from toxin-producing fungi found in moldy rice, can be converted by the host ALDH enzymes from protoxin to toxin, and the effects of mycotoxins can be further potentiated by ethanol. If the incidence of hepatic disease found in many East Asians, especially JPN, is related to the consumption of mycotoxins, then individuals carrying the deficient ALDH22 variant with lowered alcohol consumption would be selectively favored.*[1] So it sounds like it isn't alcoholism that is being selected against, but instead resistance to mouldy rice born infections, which is correlated with this gene. That explains the geographic distribution quite well, too. [1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1821113/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1821113/

by nl 5y ago

nsomaru 5y ago

If it was random, why did it become dominant if it provided no evolutionary advantage? There’s variations going on all the time, maybe they’re random, maybe they’re not. But the ones that stick in a population are the ones that get selected by outcompeting hosts without the variation.

AstralStorm 5y ago
You could probably also ask why Temujin had many children. Founder effects are a thing and a sufficient explanation. They hold even for relatively detrimental defects, and likely hold even better for something as silent as acetaldehyde metabolism.
nl 5y ago
> There’s variations going on all the time, maybe they’re random, maybe they’re not. But the ones that stick in a population are the ones that get selected by outcompeting hosts without the variation. This is only true if there is correlation between the variation and survival in terms of breeding. A good example is the high incidence of Haemophilia B in European royalty[1]. It certainly doesn't give any selective advantage, but the fact that royalty are more likely to get medical care means it doesn't have any selective disadvantage either. There are plenty of other random variations that aren't selective. Eye colour, male pattern baldness, the range of earlobe attachedness[2] are all variations that don't have any selective pressure and so their appearance in various geographic areas is down to history rather genetics. Edit: Another comment led me to a paper arguing compellingly that this gene provided some protection against diseases carried in mouldy rice, and that was selective in China. See further discussion here: https://news.ycombinator.com/item?id=27705633 https://news.ycombinator.com/item?id=27705633 [1] https://en.wikipedia.org/wiki/Haemophilia_in_European_royalty https://en.wikipedia.org/wiki/Haemophilia_in_European_royalt... [2] https://en.wikipedia.org/wiki/Earlobe#Genetics https://en.wikipedia.org/wiki/Earlobe#Genetics

dynm 5y ago

These mutations do have a lot of spread outside of China. For example Fig. 1 of this paper (http://www.medgenetics.ru/UserFile/File/Borinskaya%202009%20AJHG%20Distr%20ADH1B%2047His.pdf http://www.medgenetics.ru/UserFile/File/Borinskaya%202009%20...) shows a high incidence of ADH1B in the Middle East. Is it really required that these genes are equally selective everywhere? If the adaptation theory were true, that would depend on how widely available alcohol was, as well as how harmful it is to lose the extra calories from rotting fruit, etc. No particular reason to think it would be equal everywhere.

nl 5y ago
Thanks for this. That paper led me down a rabbit hole and I ended up with this: Evidence of positive selection on a class I ADH locus It seems unlikely that the selection was recent and associated with alcoholism, the modern phenotypic manifestation of the polymorphism.... Mycotoxins, from toxin-producing fungi found in moldy rice, can be converted by the host ALDH enzymes from protoxin to toxin, and the effects of mycotoxins can be further potentiated by ethanol. If the incidence of hepatic disease found in many East Asians, especially JPN, is related to the consumption of mycotoxins, then individuals carrying the deficient ALDH22 variant with lowered alcohol consumption would be selectively favored.*[1] So it sounds like it isn't alcoholism that is being selected against, but instead resistance to mouldy rice born infections, which is correlated with this gene. That explains the geographic distribution quite well, too. [1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1821113/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1821113/