Comment by zawerf | Hacker News Reader

Comment by zawerf | Hacker News Reader

Parent Full thread

> Completely wrong... if you're judging just between black Africans and white Europeans then it is a pretty good proxy for general genetic relatedness. It is absolutely insane to think otherwise.

Skin color is not a good proxy for genetic relatedness because it's only one of the many different traits that make up a human.

You would surely agree that two humans with different skin colors but all else equal are more closely related than two with the same skin color but nothing else in common.

And it turns out that is exactly the case. Africa has the most genetic variation since it's the origin of humans so a black African can very well be much more related to a white European than to another black African.

https://en.wikipedia.org/wiki/Human_genetic_variation

In computer science terms it would be trying to do a dimension reduction by only naively comparing the first coordinate. That would work only if skin color is the principal component of genetic diversity but it's pretty unlikely and insane to declare that a priori.

Your link directly contradicts the paragraph that precedes it.

What? I just read the link and one thing it says is that

> In general, however, an average of 85% of genetic variation exists within local populations, ~7% is between local populations within the same continent, and ~8% of variation occurs between large groups living on different continents (Lewontin 1972; Jorde et al. 2000a). The recent African origin theory for humans would predict that in Africa there exists a great deal more diversity than elsewhere and that diversity should decrease the further from Africa a population is sampled.

Which is literally what the GP said.

You should put up or not make drive-by comments.

Same quote stated more directly (emphasis mine):

> It is often stated that the fixation index for humans is about 0.15. This translates to an estimated 85% of the variation measured in the overall human population is found within individuals of the same population, and about 15% of the variation occurs between populations. These estimates imply that any two individuals from different populations are almost as likely to be more similar to each other than either is to a member of their own group

Please see my post here:

https://news.ycombinator.com/item?id=17055327

In your paragraph “similarity” refers to looking at just one genetic locus (essentially one gene) at a time.

Reply on news.ycombinator.com