3 ms·
> Is this really the relevant criterion here? I'm not sure it's the relevant criterion, but it probably is one. For instance in the original article, they tal
by grondilu 10y ago
> Is this really the relevant criterion here?
I'm not sure it's the relevant criterion, but it probably is one.
For instance in the original article, they talk about lactose tolerance and how up to 30% of the adult population currently tolerates lactose. They would not have mentioned this ratio it if it were not important.
If new mutations show up and barely keep existing without gaining any predominance, isn't it more genetic drift than evolution?
- gwern 10y agoYes. One of the signatures of evolution/selection on a gene is it spreading far faster than any random walk. For a single variant to go from 0 to up to 30% in a few thousand years is extremely fast, so that instance of human evolution was detected early on. But there are many other instances where it's a whole set of variants affecting a particular trait, all of which get slightly increased in frequency to collectively sculpt humans to their local environment; this is much harder to detect because the increase in each variant is much smaller. But it's still doable, especially with whole genomes, so look forward to more discoveries in the next few years.