I was wondering what is my most distant (human) relation on Earth? I was speculating that maybe when people left Africa about 60,000 years ago, some went off up to Europe way while others went towards Australia. So if we took an early generation age of 15 years, that would leave someone totally isolated from any of my ancestors as about my 4000th cousin.
Does that sound about right? I’m assuming anyone in Africa would be less isolated, but perhaps I’m wrong there?
Also, I read that the most recent common ancestor for anyone of European descent was surprisingly recent, at about 1000 BC. That would leave my most distant relation with European ancestry at about my 200th cousin.
Obviously these are very rough figures using only time-based guesses rather than any genetic study at all. Does anyone have any better estimates? (or even, if you can fix any flaws in my own estimates).
I think the general consensus these days is that there’s been so much mixing among populations that the odds of two people being completely isolated like that are pretty much nil.
I’m not sure if you’re asking what the real-life odds are or if you’re looking for hypothetical numbers, presupposing total isolation.
Real life numbers, if there are any. I was assuming that some Australian aboriginal people would be the least closely related to me, but if there’s a study that shows different, I’ll gladly look at that (I’m from Scotland).
I’m just trying to make an estimate, but I would like it to be as accurate as it can be.
That 1000 BC estimate–if there are any pure-blooded American Indians left it is sure to be wrong. I’m prepared to believe that most people who seem to be pure-blooded American Indians actually have at least one European or African ancestor, but all of them?
Well, there aren’t any Tasmanians left. And the Maori only colonized New Zealand in the 1300s, so they aren’t an example of an isolated population.
Australian aborigines were pretty isolated, but I’d be prepared to believe that there are no 100% full-blooded Australian aborigines left. New Guinea highlanders are another possibility. But Australia and New Guinea had a tiny amount of contact with Indonesia, and were never as isolated as the Americas. Of course, even the Americas weren’t 100% isolated, there were several waves of migration from Siberia. There are Athabaskan languages in the American southwest, which indicates some amount of travel and mixing. But it seems unlikely that Eskimo genes flowed down to South America.
Even if a tribe somewhere deep in the Amazon doesn’t have any direct contact with white folks, they probably still have contact with other, neighboring tribes, and that contact probably includes some interbreeding (maybe openly and peacefully, maybe by abducting women in warfare, maybe by star-crossed lovers meeting against their parents’ wishes, but it’ll happen). And those neighboring tribes will have contact with their neighbors, too, and so on until someone does have contact with outsiders.
I’ve wondered about the oft-heard 3000-year old MRCA claim and suspect that it’s likely false, for the reason Lemur866 and Buck Godot give. Perhaps Conquistador blood has penetrated deep into the Amazon jungle, but judging that would require sophisticated real-world modeling.
To make this point, I did a very crude model simulation where tribe A1 gets 3% of its breeders for each generation from the Conquistadors beginning 1550, tribe A2 gets 3% of its breeders from tribe A1, tribe A3 3% from tribe A2, and so on. After 20 generations (i.e., till present day), almost everyone in A5 will have some Conquistador blood, but 14% of A6 and 79% of A7 will not.
So the claim is almost certainly true if the isolation of the Amazon jungle is well described by the simple 4-deep model, and almost certainly false if the best model is 6-deep. And the model is still far too simplistic. (And perhaps there were pre-Columbian tricklings in.) In my opinion, no one knows with confidence whether the 3000-year-ago MRCA is correct or not.
Another interesting (and paradoxical?) result from population genetics is this:
You have more distinct ancestors who were alive in 1000 AD than the number who were alive in 1 AD.
Ok, I can see that extremely isolated or contactophobe groups make things difficult to figure out. I guess, other than forcibly DNA testing them, there’s no way to be sure!
Still, that leaves people in regular human society. Is 3000 years an accepted figure then? I have found that my 15 year generation is too low; looks like 20 is conservatively better.
If those are correct, then the most unrelated person I’m ever likely to meet is about my 150th cousin. Doesn’t seem that distant really! Can anyone improve on that estimate, or shall I stick with that?
Pre-european settlement, Aborigines in the north definitely traded with PNG, and even further (I have Macau in my mind, but the book I got that out of is on loan and I’m not 100% certain I’m remembering correctly there)
So, yeah, Autralia was very isolated from Europeans, but not necessarily from everywhere.
Remember we are talking about a period of at least 500 years, or about 20 generations. Even back that far, everyone has more than one million potential ancestors (assuming unique ancestors, which of course is precluded by pedigree collapse.) Over 500 years, the odds are that members of even the most remote Amazonian tribes have at least one European ancestor (and that European ancestor may have had one Mongol ancestor, etc). Going back 3,000 years, everyone has the astronomical number of 2[sup]120[/sup] potential ancestors.
Your first cousins are 2 generations removed from you. After that, each order of cousin is one more generation. That would put our common ancestor at something like 660 years ago (assuming 20 years per generation). That might be close to being true for Europeans, but unlikely to be true for the entire world.
This is of course a different problem from MRCA, as you’re seeking just pair-wise common ancestors, not a single global common ancestor. (Estimating 25 years for mother-to-child and 30 years for father-to-child is better than 20 years per generation (though best estimate is obviously culture dependent).)
A simple conservative estimate for number of cousins (which ignores the collapsing effects of shared ancestors) is that you have 2^(2k+1) k’th cousins, i.e. 8 1st cousins, 32 2nd cousins, …, so 40 million trillion 32nd cousins! Obviously the assumption of no shared ancestry isn’t working here, but the fact that the number’s saturated is indicative …
… Except that has explained upthread some remote populations may be just too isolated to be such “close” cousins.
I think the shallower model is much more similar to the actual case. And remember that none of these tribes has probably been living in exactly the same place for the past 500 years. After the Conquest, there was tremendous disruption due to mortality of up to 90% of the indigenous population from disease and other stresses, plus population movements as some groups fled into more remote areas to avoid enslavement or domination. These movements would have provided more opportunities for interbreeding, especially through consolidation of the remnants of other tribes.
In Panama, the indigenous population of the eastern part of the country was essentially wiped out by the 1600s. The area was repopulated by migrations of the Kuna and Embera peoples from Colombia in the 1700s and 1800s. Similar movements occurred elsewhere in Central and South America.
If by “most distant human relation” the OP means the human with which he or she shares the least genetic material, the answer almost certainly will be someone in sub-Saharan Africa, even if the OP himself or herself is from sub-Saharan Africa. There is more genetic variation there than in the rest of the world.
However, if by “most distant” the OP means the person as to which the closest relationship is most distant, it’s more likely to be someone in an isolated population; New Guinea may be the most plausible.