User talk:213.162.73.180

Genetic history of East Asians edit
Hello, I do not agree with your edits to the ^ page. The reason I moved Oceanians from closely related to partly related at the top section of the page is not because Oceanians form a distinct cluster from East Asians. It was because through 1-1 population sharing methods such as outgroup f3 statistics, D statistics, IBS allele sharing, and the like, Oceanians are quite distant from East Asians. Native Americans, on the other hand, do exhibit high similarity to East Asians through the aforementioned methods, and thus belong under closely related with Southeast Asians, Siberians, and Polynesians. With regard to Native Americans forming a distinct cluster from East Asians, of course they do, as they have been isolated from other human populations for something like 20,000 years. It would be ridiculous if they did not develop a distinct genetic signature since then. But you are conflating genetic distinctiveness with genetic distance. Let's say there are four East Asian individuals, A,B,C and D: C&D are siblings and A&B are strangers to each other and to C&D. Through clustering analyses such as structure or PCA, A & B would form one cluster, and C & D would form another. However, despite A & B appearing as uniform blocks of the same color in structure or clustering spatially on top of each other in PCA, they are no more closely related to each other genetically than either are to C or D. What drives the clustering is the substantial drift between C and D. That is similar to what is happening with Native Americans in clustering analyses: they are heavily bottlenecked, and so they naturally form their own pole easily, as structure looks for distinctive genetic signatures and is influenced by recent genetic drift. In terms of raw similarity, however, Native Americans are closely related to East Asians. When I kept Native Americans under the populations closely related to East Asians and moved Oceanians to partly related, I was not saying that Native Americans cannot be easily distinguished from East Asians in genetic clustering analyses or that they aren't able to form a different cluster. I was only saying that in terms of raw genetic similarity, the two are similar, which they are (more than two random West Eurasians are in fact). Paleosiberians are also closely related to East Asians. Look at the outgroup f3 statistics in Table S6 of the supplementary materials of this paper: https://www.nature.com/articles/s41431-020-0599-7, for example. Paleosiberians like Koryak, Itelmen, Chukchi, and Native Americans like Karitiana, Surui are closer to East Asians like Koreans than, say, two European populations are to each other (higher f3 value means more genetic similarity). But Paleosiberians and Native Americans also have lots of their own unique genetic drift (which you can see by their very high f3 values among themselves), which is why they are easily distinguishable from East Asians, but again, that does not mean they are genetically distant in absolute terms from them as we can see from their high f3 to them. I also do not agree with your merging of North Asians and Native Americans -- they are separate populations and I think they were better the way they were before. You rely way too heavily on PCA and structure, these are abstractions that hide huge amounts of variability within. This is just my opinion though, feel free to respond back if you disagree, I would love to hear your response and am open to discussion. Xadiiujhy (talk) 18:06, 27 January 2023 (UTC)
 * If you think so, the reason was that Native Americans are not a simple diverged East Asian group, but formed from complex admixture, yet I do not really care if you put them back, fact remains that they do not specially cluster more than let's say Melanesians do, yet having higher affinity because they derive part of their ancestry directly from Ancestral East Asians rather than a distant sibling lineage. To the second point, Native Americans cluster far closer to Northern Asians and Paleo-Siberians than they do with East Asians, which is elaborated here:, specifically: "Analyses of all 122 populations confirm many known relationships and show that most populations from North Asia form a cluster distinct from all other groups. Refinement of analyses on smaller subsets of populations reinforces the distinctiveness of North Asia and shows that the North Asia cluster identifies a region that is ancestral to Native Americans." As this is geographically structured, we may make an own section for the Americas, as there are also two for Aboriginal Australians and Melanesians, although both could (or should) be merged as simply "Oceanians". I will just restore their own paragraph, although I don't think it is necessary.213.142.97.181 (talk) 18:47, 27 January 2023 (UTC)
 * On the other hand, no, as the population history of Northern Asia is linked to the formation of Ancestral Native Americans, it therefore makes it easier to read for the readers if these informations are not split apart. Just give them a subsection.213.142.97.181 (talk) 18:50, 27 January 2023 (UTC)
 * Nobody said that Native Americans are simply an offshoot of East Asians (who themselves are a mixture of various sources), and anyway, Polynesians are not a simple diverged East Asian group either, nor are Siberians or Southeast Asians: they all have their own regional admixtures. No present day population is "pure", admixture throughout history is the norm, not exception. It's a fact that through similarity measures like outgroup f3, ibs allele sharing, d statistics, pairwise nucleotide difference, genome coalescence, etc. that native Americans have high affinity with East Asians. Clustering algorithms are not measures of genetic distance, and there are issues with PCA anyway, https://www.nature.com/articles/s41598-022-14395-4, "We demonstrate that PCA results are artifacts of the data and that they can be easily manipulated to generate desired outcomes. PCA results may not be reliable, robust, or replicable as the field assumes." Sub Saharan Africans tend to form one homogenous cluster in many studies, but genetically, two Africans are often more different from each other than either are to a European or Asian. These clustering algorithms are abstractions, and the clustering can easily change depending on the sample space, and they paint misleading portraits of discrete, homogenous populations when in reality, discrete populations do not exist. And of course, Paleosiberians are closer genetically to Native Americans than East Asians are, but that doesn't negate the fact that East Asians are also close (just not as close, obviously). Using your logic, Polynesians should not be included in closely related either, since Southeast Asians like Austronesians are closer to Polynesians than East Asians are. Also, non-paleosiberian north asians like Buryat and Ket aren't really closer to Native Americans than mainland East Asians are, it's really mostly the Paleosiberians and groups with their admixture like Ulchi or Nganasan. anyway, this is just a lighthearted conversation by the way, I was just curious as to your thought process because my own opinion is different, that's all, and it is my viewpoint at the end of the day and you are entitled to your own. Xadiiujhy (talk) 21:15, 27 January 2023 (UTC)