[ExI] digital resurrection of a genome
spike66 at att.net
Fri Jul 3 08:04:36 UTC 2015
>... On Behalf Of rex
> On Thu, Jul 2, 2015 at 6:41 PM, spike <spike66 at att.net> wrote:
> >>…Suppose a guy is married three times and has at least a couple (but as
> many as four) girlfriends on the side, and it becomes apparent that he is
> the father of at least 25 children…In theory, with sufficient persistence
> and determination, those descendants could …reconstruct the entire genome
> of the long-perished prolific ancestor...
>...About half the genes won't make it into the first successful sperm. About
1/4 won't make it into either the first or the second, and so forth.
Ja. The prolific guy was 4 generations back, so I carry about 1/16 of his genes, tragically none of the sexy ones. We have about 30 of his descendants who have done DNA testing, several of whom are his great grandchildren, so they would be carrying about 1/8 of his genes.
>...A related question: how many generations does it require for someone selected at random to have as many genes in common as a known ancestor? IOW, how many generations are required to make genealogy pointless?
About 8. Any two white* Americans are on the average about 7th cousins, so that means they share common ancestors somewhere about 8 generations back on average. If you have a cousin on any of the DNA sites and you discover a pair of commons that far back, keep looking, for there are probably other shared ancestors somewhere.
* The reason I specified white Americans is that the whole DNA game generally hasn't really caught on much in racial minorities as far as I can tell. Evidence: I can take a black cousin who can find no known white ancestors anywhere, and go into their cousin list where people have the option of putting in their photos. Those cousins tend to have plenty of white cousins, completely unexplained. Jumping to the raped slave notion doesn't get us there: three of my black cousins' ancestors came to America after the war.
The best I can do is theorize that their white cousins are far fewer but far more likely to do DNA testing. Same story with Chinese, dot Indians, feather Indians, Hispanics, Asians: their white cousins are more heavily represented by a factor of anywhere in the 3 to 8 range. Even after we theorize and compensate for some hypothetical but plausible economic factor, the asymmetry is still there and still strong. I am baffled trying to explain why. Open to suggestion.
More information about the extropy-chat